X-Risk Daily

Subscribe at buttondown.com/x-risk-daily
Thursday 18 June 2026
33 news · 4 research · 14 analysis · 9 updates from yesterday

US government orders first-ever restriction of AI model over jailbreak vulnerability

Transformative AI 17 Jun · Updated today
↻ Continues from: "White House orders halt to government AI model assessments, citing security risks"
On 12 June, the US government ordered Anthropic to disable Claude Fable 5 and Mythos 5 worldwide, just three days after Fable 5's 9 June release.
First government intervention blocking a frontier model release establishes precedent for capability-based restrictions during AI transition.

Commerce Secretary Howard Lutnick sent Anthropic CEO Dario Amodei a letter outlining the restrictions, marking the first time Washington has blocked an AI model release on national security grounds.

The directive followed warnings from Amazon researchers who flagged a jailbreak bypassing Fable's safeguards to elicit dual-use cyber capabilities. A person close to the White House told Semafor that Amazon flagged the jailbreak to the government, and that Amazon CEO Andy Jassy had been in contact with the administration about it. Fable 5 scored 53.3% on Humanity's Last Exam benchmark, compared to Claude Opus 4.8's 45.7%, and possesses capabilities similar to Claude Mythos Preview—a model Anthropic deemed too dangerous for general release in April. Mythos is understood to currently be in use by the NSA for offensive cyber operations, according to Tom's Hardware.

The export control directive required restricting access for all foreign nationals, whether inside or outside the United States, including Anthropic's own foreign-born employees. Given the scope of the directive, Anthropic argued it had no choice but to disable the models for all users. The company received the order at 5:21pm ET on 12 June and had to "abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance." Access to other Claude models, including Opus 4.8, remained unaffected.

Anthropic contested the action, arguing that the jailbreak technique "essentially consists of asking the model to read a specific codebase and fix any software flaws," and that a demonstration surfaced previously known, minor vulnerabilities also discoverable by other publicly available models, including OpenAI's GPT-5.5. The company maintained its safeguards are substantially more effective than those of any previously deployed model, and that perfect jailbreak robustness is currently impossible. Anthropic wrote: "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people." CNN reported that Anthropic argued the standard would halt all new frontier model deployments across the AI industry.

The dispute unfolded against a backdrop of prior tensions. Earlier this year, the Department of Defense declared Anthropic a supply chain risk—a designation historically applied to foreign adversaries—following the collapse of talks between the two sides. The label obligates defense contractors to certify they are not using Claude in military work. White House AI adviser David Sacks said the administration issued the export control "reluctantly" after Anthropic refused to fix the flaw or pull the model, that it wants the restriction lifted once the jailbreak is patched, and that "the ball is in Anthropic's court." The action signals Washington's willingness to invoke emergency export controls to intervene in frontier AI deployment when national security concerns emerge, setting a precedent that could reshape how American labs release models globally.

Originally from: Center for AI Safety Newsletter — Read original

US government orders Anthropic to shut down frontier models Fable 5 and Mythos 5 via export controls

Transformative AI 16 Jun · Updated today
↻ Continues from: "Proposal for frontier AI lab to voluntarily shut down to signal existential risk"
On 12 June 2026, the US Commerce Department ordered Anthropic to cut off foreign access to its most capable models, Fable 5 and Mythos 5, using export-control authority.
First use of unilateral government power to disable frontier AI system sets precedent for emergency control mechanisms during AI transition.

On 12 June 2026, the US Commerce Department ordered Anthropic to cut off foreign access to its most capable models, Fable 5 and Mythos 5, using export-control authority. Commerce Secretary Howard Lutnick sent the directive directly to Anthropic CEO Dario Amodei at 5:21pm ET, prohibiting access by any foreign national whether inside or outside the United States—including the company's own foreign-born employees.

The order marks the first known case of a commercially deployed AI model being halted through direct federal intervention. Anthropic responded by disabling both models for all customers worldwide, citing the technical and legal impossibility of filtering users by nationality in real time across cloud platforms including AWS Bedrock, Google Cloud, and Microsoft Foundry. The models had been publicly available for just three days before the shutdown. Access to all other Anthropic models remains unaffected.

The directive came ten days after the White House established a voluntary framework for pre-release review of frontier models, rather than mandatory licensing. According to Anthropic's statement, the letter provided no specific technical details of the national security concern. The company said its understanding was that the government believed it had become aware of a jailbreak technique—a method of bypassing Fable 5's safeguards designed to prevent access to the cybersecurity capabilities of the underlying Mythos model. Anthropic reviewed a demonstration and said it identified only a small number of previously known, minor vulnerabilities, and that the same jailbreak could be used on other publicly available models, including OpenAI's GPT-5.5, which are not subject to similar controls.

David Sacks, a Trump administration adviser, claimed Anthropic refused to patch the vulnerability; both this and Anthropic's account cannot be simultaneously true, but no public evidence exists to determine which is accurate. The Pentagon's chief information officer publicly supported the decision, stating the department prioritized national security over revenue cycles. The shutdown occurred with no published threshold, no technical finding, and no independent review—just a letter arriving late on a Friday afternoon. The decision represents a new instrument of state power: the ability to unilaterally disable a deployed frontier system with no transparent decision-making process, setting AI models alongside advanced semiconductors and military technology as strategically controlled assets.

Originally from: Transformer — Read original

Trump executive order requires 30-day pre-release model submission to national security agencies

Transformative AI New!
On 2 June, President Trump signed an executive order titled "Promoting Advanced Artificial Intelligence Innovation and Security", requiring AI companies to provide new AI models to the US government 30 days before general release.
Establishes mandatory pre-release government review of frontier models, shifting enforcement to national security apparatus rather than standards body.

The order directs the Secretaries of the Treasury, War (through the Director of NSA), and Homeland Security (through the director of CISA) to design a voluntary framework through which developers may submit models for evaluation. The structure represents a significant departure from earlier proposals: an earlier version gave the government up to 90 days to review advanced models before release — a timeline that was cut to 30 days in the final order after Trump worried the order would stifle American companies' lead in the global race amid competitive pressure from China.

The order distributes testing responsibility among several national security organisations including the NSA and CISA, rather than giving the Center for AI Standards and Innovation (CAISI) the central role. Reports suggest this reflected officials' push for AI national security priorities to sit within traditional security agencies rather than CAISI. Days after the order, National Cyber Director Sean Cairncross ordered CAISI to stop publishing AI model assessments, ending public transparency in frontier AI evaluation. The directive transfers oversight authority from CAISI to a classified system managed by national security agencies. CAISI had completed over 40 evaluations of AI models by early June 2026, and had announced agreements with Google DeepMind, Microsoft and Elon Musk's xAI on 5 May to evaluate their models before public release.

The timing appears linked to advances in AI capabilities for cybersecurity. Anthropic's unreleased Claude Mythos model demonstrated an extraordinary ability to autonomously detect thousands of previously overlooked high-severity zero-day vulnerabilities within major operating systems. According to CNN, Mythos sparked concerns among governments, banks and utility companies, with Anthropic restricting access to approved organisations rather than releasing the model publicly. The Mythos announcement came in April, one month before the CAISI partnerships were formalised and weeks before the executive order — suggesting the model's capabilities may have accelerated government action.

The shift toward classified evaluation has implications for transparency and competition. CAISI's public evaluations served as a kind of neutral benchmarking service; with evaluations now classified, that independent verification disappears. US-based AI firms are now subject to a review process that Chinese, European, and other international competitors are not. The move represents what Scientific American described as a fundamental shift from the administration's previous hands-off approach to the technology, reflecting how the development of more powerful AI models has spooked some federal officials, prompting the White House to reverse course and back some safety measures.

Originally from: Center for AI Safety Newsletter — Read original

NSA reportedly using Anthropic's Mythos model for offensive cyber operations

Transformative AI New!
The National Security Agency is using Anthropic's Claude Mythos model for offensive cyber operations, according to a Financial Times report that marks the first confirmed deployment of frontier AI capabilities for government cyberwarfare.
Government deployment of AI for offensive cyber operations demonstrates willingness to weaponise frontier capabilities during AI transition.

The National Security Agency is using Anthropic's Claude Mythos model for offensive cyber operations, according to a Financial Times report that marks the first confirmed deployment of frontier AI capabilities for government cyberwarfare. The arrangement is particularly striking given that the Department of Defense designated Anthropic a "supply chain risk" earlier this year, effectively blacklisting the company from federal contracts.

Anthropic has embedded approximately six forward-deployed engineers inside the NSA to guide the agency's use of Mythos and customize the model for specialized applications, according to Tom's Hardware. Sources told the Financial Times that Mythos could be used to infiltrate the networks of other states, notably China and Iran. Mythos is the version of Anthropic's most capable model that the company deemed too dangerous for public release due to its cyber vulnerability exploitation capabilities, with Anthropic stating it can identify and exploit zero-day vulnerabilities in every major operating system and web browser.

The collaboration represents a sharp contradiction in the government's posture toward Anthropic. The dispute between Anthropic and the Pentagon began in January 2026, when the two parties were negotiating a $200 million contract and the Trump administration demanded that Anthropic allow usage of its technology for "all lawful purposes," implying the removal of AI guardrails — a move that conflicted with the company's usage policy. After Anthropic withdrew from that contract over concerns about domestic surveillance and autonomous weaponry, the Pentagon signed agreements with OpenAI, Google, and xAI instead. Yet the NSA sits under the Department of Defense, the same department arguing in court that Anthropic's technology poses a national security risk, though reports suggest the NSA was already using Mythos despite the blacklist, according to TechSpot.

The revelation raises fundamental questions about the dual nature of AI safety work — models withheld from public release due to danger are being provided to government agencies for exactly the capabilities deemed too risky for general availability. This pattern extends beyond Anthropic: the US government's restriction of Fable 5 over concerns about jailbroken cyber capabilities suggests a policy of government monopoly on dangerous cyber AI rather than preventing development of such capabilities entirely. Anthropic has framed Mythos as a defensive cybersecurity tool, launching Project Glasswing in April with partners including AWS, Google, Microsoft, Nvidia, and CrowdStrike, and this week announced that partners had found more than 10,000 high- or critical-severity flaws, with access expanding to approximately 150 organizations across 15 countries.

Originally from: Center for AI Safety Newsletter — Read original

US and Iran sign nuclear agreement with $300bn redevelopment package

Geopolitics & Conflict New!
On 18 June 2026, the United States and Iran reached a comprehensive agreement that includes an end to hostilities, Iran's commitment never to develop nuclear weapons, and a $300 billion economic redevelopment package for Iran.
Nuclear non-proliferation and reduction of Middle East conflict risk involving nuclear-capable powers.
The 14-paragraph memorandum represents a major diplomatic breakthrough in one of the world's most persistent nuclear proliferation crises. The agreement's nuclear provisions directly address a key pathway to catastrophic conflict in the Middle East — a region where nuclear escalation could draw in multiple great powers. The economic component suggests substantial American commitment to making the deal durable, potentially reducing Iran's incentives to pursue covert weapons programmes. However, the brief report provides no detail on verification mechanisms, enforcement provisions, or how the agreement addresses Iran's existing nuclear infrastructure and expertise. Previous nuclear agreements with Iran have faced implementation challenges, and the durability of this framework will depend heavily on technical details not yet disclosed.
Source: BBC News - World — Read original
Transformative AI

SpaceX goes public at $2.5 trillion valuation, acquires Cursor for $60 billion

Transformative AI New!
On 12 June, SpaceX — the parent company of xAI — went public and reached a valuation of over $2.5 trillion.
Major capital concentration in AI development tools controlled by single actor with significant influence over AI trajectory and governance.
Four days later, on 16 June, it exercised its option to purchase Anysphere, the creators of the AI coding assistant Cursor, for $60 billion. The acquisition gives xAI control of one of the leading AI-powered development tools at a time when automated coding is becoming central to AI research acceleration. The $60 billion price tag for a coding tool company reflects the strategic importance companies are placing on AI development automation. Combined with SpaceX's massive valuation, this represents significant capital concentration in Elon Musk's companies during the critical period of AI development that Anthropic has described as potentially leading to recursive self-improvement.
Source: Center for AI Safety Newsletter — Read original

DeepSeek projected to raise $7.4 billion in first funding round

Transformative AI New!
Chinese AI company DeepSeek is projected to raise $7.4 billion in its first funding round, representing substantial investment in a major Chinese AI developer.
Significant capital flows to Chinese frontier AI development maintain competitive pressure during critical transition period.
The funding round indicates continued strong capital flows to Chinese frontier AI development despite geopolitical tensions and export controls. DeepSeek has been developing competitive models and the capital injection will likely accelerate its research capabilities. The substantial funding reflects both China's commitment to AI development and investor confidence in Chinese AI companies' ability to compete with Western labs despite technology restrictions.
Source: Center for AI Safety Newsletter — Read original

Representatives release draft bill requiring independent audits of frontier AI developers

Transformative AI New!
Representatives Jay Obernolte and Lori Trahan released a draft of the Great American AI Act, including proposals for mandatory independent audits of frontier AI developers.
Proposed federal legislation would mandate independent safety audits of frontier AI developers with uniform national standards.
The draft legislation includes a federal preemption clause that would override local laws on AI development while preserving local regulations on AI deployment. Mandatory independent audits would represent a significant increase in external oversight of frontier labs' safety practices and capability evaluations. The federal preemption aspect suggests an attempt to create uniform national standards for AI development while allowing variation in how AI systems are used locally. The distinction between development and deployment regulation indicates recognition that frontier AI development poses different governance challenges than AI application.
Source: Center for AI Safety Newsletter — Read original

Former UK AI Safety Institute researchers launch Sequent, aiming for $100-150M to pursue differentiated alignment research

Transformative AI 15 Jun · Updated today
↻ Continues from: "AI safety researchers launch Sequent, aiming for 40-80 staff and theoretical guarantees on alignment"
On 10 June, senior AI safety researchers announced Sequent, a new nonprofit alignment research organisation targeting $100-150 million in initial funding and 40-80 full-time researchers within two years.
Directly addresses the core alignment problem during the transition to superintelligence — credible researchers taking costly action based on inside knowledge.

On 10 June, senior AI safety researchers announced Sequent, a new nonprofit alignment research organisation targeting $100-150 million in initial funding and 40-80 full-time researchers within two years. Led by Geoffrey Irving, formerly Chief Scientist at the UK AI Safety Institute and previously at DeepMind, OpenAI, and Google Brain, alongside Daniel Murfet from Timaeus, the organisation represents a significant bet on theory-driven approaches to artificial superintelligence alignment.

Sequent's central thesis is that empirical programmes at major AI labs are unlikely to deliver high prior confidence that superintelligent systems will behave as intended. The organisation aims instead to pursue what it calls a portfolio of theoretical and empirical bets that, if any succeed, would provide stronger a priori guarantees before training advanced AI systems. Research areas include scalable oversight techniques such as debate and amplification — methods Irving helped pioneer during his tenure at OpenAI — as well as singular learning theory, heuristic arguments, and game-theoretic frameworks. The organisation plans heavy investment in automated research tools, arguing that theoretical approaches offer better filters for determining which automated directions hold promise.

To preserve the advantages of smaller alignment teams — research focus, opinionated leadership, and low coordination overhead — Sequent will adopt a federated structure in which a handful of research directors maintain substantial autonomy over research direction, team culture, and hiring within their areas. These directors will report to Irving, and the final portfolio of research areas will depend on which senior researchers join. The organisation explicitly seeks to remain independent rather than join an existing AI lab, citing the need to maintain the freedom to raise concerns if fundamental obstacles emerge and to avoid institutional pressure toward purely empirical approaches.

The launch comes at a moment of growing concern about whether alignment research will keep pace with capabilities development. Sequent acknowledges it may exacerbate the bottleneck of experienced alignment researchers available to other efforts, but contends that no comparable large-scale theory-focused organisation currently exists. Whether automated alignment research can deliver theoretical guarantees before the arrival of transformative AI systems remains an open question, one that Sequent's substantial funding target suggests will require both significant resources and a departure from current laboratory norms.

Go deeper: Sequent announcement on Alignment Forum

Originally from: Import AI — Read original

Anthropic expands Mythos access to 150 additional organisations through Project Glasswing

Transformative AI New!
Anthropic expanded Project Glasswing, extending Claude Mythos access to approximately 150 more organisations.
Broader controlled distribution of unrestricted frontier model expands number of actors with access to dangerous capabilities.
Mythos is the version of Anthropic's most capable model without strict bio or cyber safeguards, which the company has deemed too dangerous for public release. Project Glasswing provides controlled access to Mythos for trusted organisations, presumably for research, evaluation, or specific use cases where the additional capabilities are needed. The expansion to 150 additional organisations represents a significant broadening of access to Anthropic's most capable and potentially dangerous model, even as the company restricts its standard Fable model. This creates a two-tier system where selected organisations can access capabilities deemed too risky for general availability.
Source: Center for AI Safety Newsletter — Read original

Congressional AI export control bills gain bipartisan momentum as White House regulatory approach falters

Transformative AI 16 Jun
The Republican-controlled House Foreign Affairs Committee has approved eighteen export control bills in recent months, the largest such legislative package in history, with several measures expected to be incorporated into the 2026 National Defense Authorization Act.
Congressional moves toward enforceable AI chip controls — could restore export discipline and extend US compute advantage during AI transition.

The surge in congressional activity reflects deepening frustration with executive inaction on technology controls targeting China's semiconductor and AI capabilities.

The most consequential piece of legislation is the Multilateral Alignment of Technology Controls on Hardware (MATCH) Act, introduced by Representative Michael Baumgartner in early April. The bill would compel allied nations to impose export controls on advanced chipmaking equipment sales to China equivalent to those maintained by the United States, threatening to invoke the Foreign Direct Product Rule if allies fail to harmonize their restrictions. China's imports of semiconductor manufacturing equipment surged from $10.7 billion in 2016 to approximately $51.1 billion in 2025, according to analysis from Silverado Policy Accelerator, highlighting the scale of the challenge. The legislation targets a critical asymmetry: while U.S. companies face stringent controls, allied firms from the Netherlands, Japan, and South Korea have continued servicing and selling equipment to Chinese customers, allowing Beijing to stockpile chokepoint technologies like deep ultraviolet lithography machines.

Equally significant is the AI Overwatch Act, which the House Foreign Affairs Committee advanced on 21 January by a vote of 42-2. The bill would impose a statutory two-year ban on exports of Nvidia's Blackwell-class chips to China and require the Commerce Department to notify Congress before approving licenses for advanced AI chip exports to designated high-risk countries, granting lawmakers the power to block transactions through a joint resolution of disapproval. This arms-sale-style oversight mechanism represents a direct congressional challenge to executive control over technology policy, coming in the wake of the Trump administration's decision to shift H200 chip exports from presumption of denial to case-by-case review in January 2026.

The congressional push reflects a broader pattern: the Trump administration has imposed no new technology-based controls on China since taking office, while enforcement gaps—including a loophole that allowed Chinese subsidiaries to purchase advanced AI chips—went unaddressed for over a year. Allied governments report confusion about U.S. strategy, with the executive branch signaling openness to commerce with China while Congress advances restrictive legislation. This discord creates negotiating leverage: statutory restrictions would allow the administration to position controls as beyond its discretion when engaging with Beijing and allied capitals. Sources indicate Chinese officials are lobbying heavily against the MATCH Act, suggesting genuine concern about its potential to disrupt China's semiconductor indigenization efforts. Whether the NDAA ultimately includes symbolic gestures or substantive measures like MATCH and AI Overwatch will determine whether congressional hawks succeed in reclaiming control over China technology policy from an executive branch perceived as prioritizing diplomatic stability over technological containment.

Originally from: ChinaTalk — Read original

France to replace Palantir AI tools with domestic provider to avoid US dependency

Transformative AI 16 Jun
France's domestic intelligence service is ending its use of Palantir's AI data tools in favour of domestic provider ChapsVision, Prime Minister Sébastien Lecornu announced on 16 June.
Reflects fragmentation of international AI cooperation and concentration of AI capabilities along geopolitical lines during the transformative AI transition.
The decision is driven by concerns about "strategic dependency" on US-controlled AI systems in critical national security infrastructure. Lecornu stated that France "cannot accept new strategic dependencies in the digital sphere" and must develop its own AI capabilities rather than relying on tools from foreign powers. The move reflects growing European anxiety about dependence on American technology companies for sensitive government functions, particularly as AI systems become more deeply embedded in intelligence operations. While the immediate switch is to a French provider, the announcement signals a broader policy shift toward technological sovereignty in AI deployment. The decision comes amid wider debates about compute governance, access to frontier AI systems, and the geopolitical concentration of AI development capabilities in a small number of US-based companies.
Source: The Guardian — Read original

Elon Musk becomes world's first trillionaire as SpaceX debuts at $2.2tn valuation

Transformative AI 12 Jun · Updated today
↻ Continues from: "Elon Musk becomes world's first trillionaire as SpaceX debuts at $2.2tn valuation"
Elon Musk's net worth reached $1.11 trillion on 12 June following SpaceX's stock market debut on the Nasdaq, with the company valued at $2.2 trillion, according to Bloomberg.
Power concentration—unprecedented wealth in the hands of a figure with direct control over frontier AI development and stated scepticism of external safety oversight.
The listing represents a significant concentration of wealth and influence in the hands of a figure who controls multiple strategically important companies, including xAI, Tesla, and Neuralink, alongside SpaceX. Musk has previously expressed views on AI development that diverge from mainstream safety perspectives and has demonstrated willingness to pursue AI capabilities development with limited external oversight. The extreme wealth concentration—Musk's fortune now exceeds the GDP of most nations—potentially amplifies his ability to shape the trajectory of transformative AI development through xAI and influence related policy debates. The SpaceX valuation itself reflects the company's dominance in satellite deployment, which has implications for AI compute infrastructure and global communications networks during the AI transition.
Source: BBC News - Science & Environment — Read original

Cognition releases FrontierCode benchmark; Claude Opus 4.8 achieves only 13.4% on hardest tier

Transformative AI 15 Jun
On 8 June, Cognition released FrontierCode, a coding benchmark designed to measure whether AI-generated code meets the standards human maintainers would accept in production, rather than merely testing functional correctness.
High-quality evaluation infrastructure for tracking capability progress toward autonomous software development — a key step toward recursive self-improvement.

On 8 June, Cognition released FrontierCode, a coding benchmark designed to measure whether AI-generated code meets the standards human maintainers would accept in production, rather than merely testing functional correctness. The benchmark comprises 150 hand-crafted tasks spanning Python, Go, TypeScript, JavaScript, Java, C/C++, and other languages, with each task requiring more than 40 hours of work by leading open-source developers. Tasks are evaluated across six dimensions — correctness, test quality, scope discipline, style adherence, maintainability, and regression safety — using a grading system in which any "blocker" issue earns an automatic zero, even if other aspects of the code are sound.

On the hardest Diamond tier, which contains 50 tasks, Claude Opus 4.8 achieved only 13.4%, followed by GPT-5.5 at 6.3% and Claude Opus 4.7 at 5.2%. Performance improved on the Main tier (100 tasks including Diamond) to 34.3%, 25.5%, and 23% respectively, and on the Extended tier (all 150 tasks) to 51.8%, 44.8%, and 43.2%. The low scores reflect a gap between code that runs and code that satisfies the discipline expected in professional codebases — what Cognition describes as the difference between passing unit tests and earning approval from a repository maintainer.

The benchmark's difficulty stands in sharp contrast to earlier evaluations. SWE-Bench, introduced in October 2023, has shown signs of saturation, with leading models now scoring above 50% on many variants. Cognition's initiative aims to establish a new standard for what it terms "maintainable code," positioning FrontierCode as the third era of AI coding benchmarks after autocomplete (HumanEval, 2021) and test-passing (SWE-Bench, 2023). The company has opened evaluation to all model creators, framing the benchmark as a measure of production readiness for autonomous coding agents.

FrontierCode's focus on mergeability addresses what some researchers view as a systemic weakness in current coding agents. Tasks assess not only whether code produces correct output, but whether it introduces unnecessary scope changes, maintains consistent style, includes appropriate tests, and avoids subtle antipatterns — criteria that are difficult to encode in binary pass-fail tests. One example task involved refactoring warning logs into a new function; Claude Opus 4.8 produced functionally equivalent code but mixed logging patterns in ways that would complicate future maintenance, illustrating the nuanced quality gaps the benchmark is designed to capture.

The release comes amid rapid iteration cycles among frontier labs. Claude Opus 4.8 was released on 28 May 2026, just 41 days after its predecessor. A subsequent model, Claude Fable 5, launched in mid-June and more than doubled the Diamond score to 29.3%, suggesting the benchmark may saturate faster than Cognition anticipated — though scores remain well below the thresholds seen on earlier evaluations, and the low baseline reinforces the view that production-grade agentic coding remains an unsolved problem.

Originally from: Import AI — Read original

Germany establishes AI Security Institute modelled on UK's AISI

Transformative AI 15 Jun
Germany's National Security Council decided to establish a national AI Security Institute based on the UK's model.
AI governance infrastructure — another major power establishes dedicated safety evaluation capacity.
The announcement represents an expansion of government-led AI safety evaluation infrastructure among major economies. No details were provided about timeline, staffing, or the institute's specific mandate. The decision follows the UK AISI's establishment and comes amid growing international focus on frontier AI evaluation capabilities.
Source: Sentinel Global Risks Watch — Read original

Xiaomi releases 1T-parameter model generating 1000 tokens per second on commodity hardware

Transformative AI 15 Jun
Chinese technology company Xiaomi published details on 15 June of MiMo-V2.5-Pro-UltraSpeed, a 1 trillion parameter language model capable of generating 1000 tokens per second on an 8-GPU commodity node.
Demonstrates continued capability progress in inference efficiency, potentially enabling faster iteration cycles for autonomous AI development.
The system achieves this speed through co-design of the model and inference stack, including FP4 quantization, DFlash (a speculative decoding method based on block-level masked parallel prediction), and close integration with TileRT software from startup Tile AI. Xiaomi emphasises that the model runs on commodity hardware rather than specialised infrastructure. The company positions the work as unlocking novel capabilities — such as rapid real-time software refactoring — that become possible when generation speed crosses certain thresholds. The development also reflects a broader trend among Chinese companies to maximise performance and efficiency from AI systems, potentially in response to export controls limiting access to more performant hardware.
Source: Import AI — Read original

OpenAI outlines goal to build automated AI researcher by March 2028

Transformative AI 12 Jun
On 28 October 2025, OpenAI CEO Sam Altman and chief scientist Jakub Pachocki announced during a livestream that the company is targeting March 2028 to build a fully autonomous AI researcher—a system capable of running independent research projects from conception to completion.
Explicit 2028 timeline for automated AI researcher from OpenAI leadership reveals expectations about recursive self-improvement and transformative AI arrival.

On 28 October 2025, OpenAI CEO Sam Altman and chief scientist Jakub Pachocki announced during a livestream that the company is targeting March 2028 to build a fully autonomous AI researcher—a system capable of running independent research projects from conception to completion. The announcement laid out three core goals: building an automated AI researcher that remains steerable and accountable, accelerating the economy through scientific progress, and delivering personal AGI to everyone on Earth.

The timeline includes an intermediate milestone: an AI research intern by September 2026, designed to meaningfully accelerate human scientific work. According to The Decoder, Pachocki emphasized that the research intern would significantly speed up OpenAI's own researchers, while the March 2028 system would handle entire research workflows autonomously. The explicit less-than-two-year timeframe from mid-2026 to early 2028 represents OpenAI's most concrete public statement about when it expects to achieve systems capable of recursive self-improvement—a threshold widely considered pivotal in discussions of transformative AI risk.

Pachocki outlined the technical foundations underpinning these ambitions, pointing to continued scaling of deep learning systems and advances in "in-context compute"—runtime processing power that extends a model's reasoning capacity. The Decoder reported that OpenAI plans to dramatically extend the time horizons over which models can reason, moving well beyond current capabilities. Pachocki also introduced a five-layer safety model spanning value alignment, goal alignment, reliability, adversarial robustness, and systemic safety, with Chain-of-Thought Faithfulness emerging as a central research area to manage portions of internal reasoning that may remain unsupervised.

The announcement arrived the same day OpenAI finalized its restructuring into a public benefit corporation, separating from its original non-profit charter. The March 2028 target aligns with statements from OpenAI co-founder Greg Brockman, who said he expects AGI within one to three years and that he would consider it a failure if the company had not reached AGI by 2030, according to Prinz AI. During the livestream, Altman emphasized that defining a concrete target—an automated AI researcher—was more useful than attempting to satisfy varied interpretations of AGI. The framing of universal personal AGI as a top-level corporate goal signals OpenAI's vision for post-AGI deployment, though the company has provided no detail on distribution mechanisms or timelines beyond the research automation milestone.

Originally from: Transformer — Read original

Senate Armed Services Committee Approves AI Guardrails Act for Pentagon

Transformative AI 12 Jun
On 12 June, the Senate Armed Services Committee incorporated Senator Elissa Slotkin's AI Guardrails Act into the National Defense Authorization Act markup, establishing what Slotkin described as the first statutory constraints on Pentagon AI use, particularly for life-and-death decisions.
Establishes legislative constraints on military AI deployment, particularly autonomous weapons — directly addresses AI-enabled catastrophic risks in military contexts.

On 12 June, the Senate Armed Services Committee incorporated Senator Elissa Slotkin's AI Guardrails Act into the National Defense Authorization Act markup, establishing what Slotkin described as the first statutory constraints on Pentagon AI use, particularly for life-and-death decisions. The legislation mandates that human beings remain the ultimate decision makers in the kill chain, with specific prohibitions on AI making final decisions on nuclear weapon deployment, domestic surveillance, or lethal targeting without human oversight.

Slotkin, who introduced the standalone bill in March, argued that no single Secretary of Defense or AI company should unilaterally set rules for AI weapons deployment — such decisions should be legislated to prevent arbitrary changes by future administrations. The provision also mandates rigorous testing of AI systems before deployment, applying standards comparable to or exceeding those used for traditional weapons systems. The move comes amid heightened congressional interest in military AI governance, with fellow Armed Services Committee member Senator Kirsten Gillibrand also introducing parallel legislation, the Secure and Accountable Military AI Act, which would impose similar restrictions on AI use for launching nuclear weapons, surveilling Americans, and developing autonomous weapon systems.

The legislative push follows a public dispute between the Pentagon and AI firm Anthropic, which culminated in the Department of Defense designating Anthropic a supply chain risk and severing contracts after the company pressed for specific assurances around autonomous weapons and mass surveillance. Slotkin's legislation appears designed to codify the type of guardrails Anthropic had sought, framing them as essential safeguards rather than obstacles to AI adoption. She has emphasised that the guardrails align with the Trump administration's AI Action Plan, which calls for aggressive AI adoption by the armed forces while ensuring systems are secure and reliable.

Beyond military applications, Slotkin is separately working on legislation to prevent AI from making final decisions on veterans' healthcare benefits, allowing AI only as a decision support tool. The NDAA markup occurred behind closed doors, which Slotkin credits for enabling substantive bipartisan negotiation on AI constraints — a rare area of cross-party agreement in an otherwise fractious policy landscape. The full text of the AI Guardrails Act, available through Congress.gov, runs to just five pages and establishes what supporters describe as left and right limits on Pentagon AI deployment without impeding technological competitiveness against adversaries such as China.

Originally from: ChinaTalk — Read original
Geopolitics & Conflict

Trump announces US-Iran peace agreement at Versailles, opening Strait of Hormuz

Geopolitics & Conflict New!
On 17 June 2026, President Donald Trump told reporters at Versailles that he had signed a peace deal with Iran, announcing the immediate opening of the Strait of Hormuz.
Major de-escalation between US and Iran could reduce nuclear weapons risk and regional instability during the AI transition.
The agreement, described as a memorandum of understanding, establishes a framework under which Iran would cease funding terrorism and abandon nuclear weapons development in exchange for reintegration into the global economy. Trump denied reports that the deal includes a $300 billion fund for Iran or commitments from Gulf states, stating the US would not contribute financially and that other nations could invest if they chose. The announcement came during a G7 meeting coinciding with celebrations of 250 years of American independence. If implemented, the agreement could represent a major shift in Middle Eastern geopolitics and nuclear proliferation risk, though the durability of any US-Iran framework has historically been uncertain. The deal's verification mechanisms and enforcement provisions were not detailed in the initial announcement.
Source: The Guardian — Read original

Iranian oil tankers breach US naval blockade in Gulf of Oman

Geopolitics & Conflict New!
On 17 June, three Iranian tankers carrying crude oil successfully passed through a US naval blockade line in the Gulf of Oman, according to ship-tracking data reported by BBC News.
Great-power instability and potential military escalation in a region critical to global energy security.
The breach represents a significant escalation in US-Iran tensions, as it demonstrates Iran's willingness to challenge American military enforcement directly. The incident raises questions about the effectiveness of US naval operations in the region and could embolden further Iranian defiance of Western sanctions and military pressure. The blockade itself appears to be part of broader US efforts to restrict Iranian oil exports, though the article provides limited detail on the blockade's legal basis or operational parameters. The successful passage of the tankers may increase the likelihood of military confrontation between US and Iranian forces in the strategically vital Strait of Hormuz region, through which approximately one-fifth of global oil supplies transit. Whether this incident prompts a US military response or signals a shift in American enforcement posture remains unclear. The development occurs against a backdrop of long-standing nuclear tensions between Iran and Western powers.
Source: BBC News - World — Read original

Japan's Defence Minister calls for end to post-war pacifism as regional tensions rise

Geopolitics & Conflict New!
Japanese Defence Minister Shinjiro Koizumi told the BBC that Japan must abandon its post-World War Two pacifist stance, arguing that ramping up defence capabilities is 'critical' to preventing conflict in the region.
Regional military posture changes affecting great-power stability during the AI transition.
The statement marks a significant shift in rhetoric from a country whose constitution has severely restricted military activity since 1945. Koizumi's comments come amid rising tensions with China over Taiwan and North Korea's expanding nuclear programme. Japan has been gradually increasing defence spending in recent years, with Prime Minister Fumio Kishida's government pledging to double the defence budget to 2% of GDP by 2027. The call to 'revisit' pacifism suggests a potential constitutional reinterpretation or amendment, which would represent a major departure from Japan's security doctrine. The shift reflects broader regional instability and the weakening of US security guarantees under recent administrations. Japan's military posture matters for global stability: as the world's third-largest economy and a key US ally in East Asia, its defence policy affects the balance of power in a region where miscalculation between nuclear-armed states could escalate rapidly during the AI transition.
Source: BBC News - World — Read original

Strait of Hormuz remains largely closed despite ceasefire, threatening global oil supply

Geopolitics & Conflict 16 Jun · Updated today
↻ Continues from: "US-Iran ceasefire ends brief Strait of Hormuz conflict with thousands dead, regional order unchanged"
Shipping through the Strait of Hormuz — the critical chokepoint handling roughly 21% of global oil supply — remains severely restricted more than a month after a ceasefire ended direct conflict in the region.
Prolonged Strait of Hormuz closure threatens global energy security and economic stability during AI transition period; potential catalyst for great-power conflict.
Three key obstacles prevent a return to normal traffic levels, according to maritime security experts interviewed by the BBC. First, the strait remains heavily mined from the conflict, with no coordinated demining effort yet underway. Second, insurance costs for vessels transiting the area have risen by orders of magnitude, making commercial passage economically unviable for most operators. Third, Iran has imposed new transit tolls and inspection requirements that shipping companies view as prohibitive. The disruption has already pushed Brent crude above $140 per barrel, the highest level since 2022. Energy analysts warn that if the strait does not reopen to substantial traffic within three months, European economies could face fuel rationing and industrial shutdowns by autumn. The impasse reflects deeper geopolitical tensions, as Western powers resist paying Iranian tolls while lacking the naval capacity to guarantee safe passage through contested waters still controlled by Tehran's Revolutionary Guard.
Source: BBC News - World — Read original

BBC investigation reveals Russian intelligence directed arson plots targeting UK Prime Minister

Geopolitics & Conflict 15 Jun · Updated today
↻ Continues from: "BBC investigation reveals Russian intelligence directed arson attacks targeting UK Prime Minister"
A BBC investigation has uncovered evidence that Russian intelligence services orchestrated arson attacks targeting the UK Prime Minister, disclosed on 15 June 2026.
Direct escalation of state-sponsored violence against democratic leadership during geopolitical crisis — potential catalyst for NATO-Russia confrontation.
The operation involved not only direct sabotage attempts but also a coordinated disinformation campaign using fabricated far-right and Muslim group identities to inflame domestic tensions. The evidence suggests Russian services are actively attempting to destabilise a NATO member state through both kinetic attacks on senior government figures and information operations designed to exacerbate social divisions. This represents an escalation from cyber operations and influence campaigns to direct physical threats against Western democratic leadership. The targeting of a sitting prime minister, combined with simultaneous efforts to manufacture sectarian conflict, indicates Russian willingness to take significant risks during a period of heightened geopolitical tension. UK security services have reportedly been briefed on the findings, though the full scope of the plot and whether arrests have been made remains unclear. The incident raises questions about the adequacy of current countermeasures against state-sponsored sabotage operations in NATO countries.
Source: BBC News - Europe — Read original

US lifts naval blockade of Iran, IAEA inspectors to return under new agreement

Geopolitics & Conflict 16 Jun
The United States has lifted its naval blockade of the Strait of Hormuz and signed a memorandum of understanding with Iran that includes the return of International Atomic Energy Agency inspectors, US Vice President JD Vance announced on 16 June 2026.
Major de-escalation of US-Iran confrontation reduces immediate nuclear escalation risk and restores monitoring of Iranian nuclear programme.
Iranian vessels have begun passing through the strait following the lifting of restrictions. The deal has triggered a backlash in Israel, which views renewed IAEA access to Iranian nuclear facilities with concern. The agreement represents a significant de-escalation in a standoff that had threatened global oil supplies and risked military confrontation between the US and Iran. The return of inspectors suggests some form of nuclear monitoring framework is being re-established, though the full terms of the memorandum have not been disclosed. The Israeli reaction indicates potential fractures in US-Israel coordination on Iran policy during a period when preventing Iranian nuclear weapons development remains a stated priority for both countries.
Source: Al Jazeera English — Read original

Trump-Iran ceasefire deal leaves Netanyahu in political bind

Geopolitics & Conflict 16 Jun
US President Donald Trump has brokered a ceasefire agreement with Iran, creating a significant political and security challenge for Israeli Prime Minister Benjamin Netanyahu.
Shifts regional power dynamics during a period of nuclear proliferation risk and geopolitical instability in the Middle East.
The deal, announced on 16 June 2026, represents a major shift in Middle East dynamics and potentially constrains Israel's strategic options regarding Iran's nuclear programme and regional influence. Netanyahu now faces pressure to accept a diplomatic framework negotiated without Israeli input, while hardliners in his coalition government may view any accommodation with Iran as unacceptable. The agreement also signals a potential realignment of US priorities in the region, with Trump prioritising direct bilateral engagement with Tehran over coordination with traditional allies. The ceasefire's terms and durability remain unclear, but its immediate effect is to complicate Israel's security posture during a period of ongoing regional tensions. This development could affect the stability of Netanyahu's governing coalition and Israel's ability to act unilaterally against perceived Iranian threats.
Source: BBC News - World — Read original

US-Iran agreement faces Republican scepticism as Vance says details remain unresolved

Geopolitics & Conflict 16 Jun
On 16 June, Vice-President JD Vance acknowledged that significant details of a US-Iran agreement announced earlier this week remain to be finalised, as Senate Republicans questioned the deal and demanded fuller disclosure from the White House.
Potential de-escalation in US-Iran tensions could reduce nuclear risk and great-power instability in a critical strategic region.
The memorandum of understanding, announced on Sunday and scheduled for ceremonial signing on Friday in Geneva, centres on reopening the Strait of Hormuz and lifting the US naval blockade in the region. The agreement includes financial incentives for Iran contingent on meeting unspecified benchmarks. Republicans have expressed particular concern about the inclusion of funds for Iran and have pressed for clarity on what conditions Tehran must fulfil. The deal is framed as ending "the war in Iran" — though the nature of this conflict is not specified in the available reporting. The agreement represents a potential de-escalation in US-Iran tensions, though the lack of detail and internal Republican opposition suggest implementation remains uncertain.
Source: The Guardian — Read original

Netanyahu declares indefinite occupation of Lebanon, Gaza, and Syria as 'security zones'

Geopolitics & Conflict 15 Jun
On 15 June 2026, Israeli Prime Minister Benjamin Netanyahu announced that Israeli forces would maintain indefinite occupation of what he termed "deep security zones" in Lebanon, Gaza, and Syria.
Regional destabilisation in a nuclear-armed part of the world during the AI transition; potential fragmentation of international cooperation.
In a televised press conference, Netanyahu declared a "historic victory over Iran" and ruled out any immediate withdrawal from Lebanese territory, stating Israeli forces would remain "for as long as necessary." The announcement followed a preliminary agreement between Washington and Tehran, which has provoked anger within Israel and drawn criticism of Netanyahu's government. The statement represents a significant escalation in Israel's territorial posture, moving from temporary military operations to announced long-term occupation of neighbouring states. This marks a substantial shift in Middle Eastern geopolitics, with potential implications for regional stability, US-Iran relations, and the broader security architecture during a period when international cooperation on existential risk management may be critical.
Source: The Guardian — Read original

Republicans attack Trump-Iran deal despite administration's claims of major victory

Geopolitics & Conflict 17 Jun · Updated today
↻ Continues from: "Iran frames US nuclear deal as victory despite domestic economic pressures"
On 17 June 2026, Republican Senator Bill Cassidy denounced the Trump administration's newly released interim agreement with Iran as "the worst foreign policy blunder in decades," drawing comparisons to Reagan-era foreign policy to criticise the deal.
Geopolitical stabilisation in the Gulf region reduces near-term economic disruption risk, though concessions to Iran may affect nuclear proliferation dynamics.
The 14-point accord, made public on Wednesday, ends a 110-day conflict and aims to reopen the Strait of Hormuz — a critical global shipping chokepoint whose closure threatened worldwide economic disruption. While the Trump administration framed the agreement as a "major win," the deal reportedly includes significant political and financial concessions to Tehran. The partisan split is notable: criticism is coming from within Trump's own party, suggesting substantial concern over the terms negotiated. The conflict's resolution reduces immediate risk of economic collapse from blocked oil shipments, but the concessions raise questions about what constraints on Iran's nuclear programme or regional activities may have been traded away. The article does not specify what concessions were made or whether the deal addresses Iran's nuclear capabilities.
Source: The Guardian — Read original
Biosecurity

AI CEOs sign open letter calling for DNA synthesis screening to prevent bioweapons

Biosecurity New!
AI company CEOs signed an open letter calling for screening of synthetic DNA orders to prevent malicious actors from obtaining AI-designed bioweapons.
Industry acknowledgment that AI-assisted bioweapon design is becoming feasible, requiring infrastructure-level safeguards.
The letter represents industry acknowledgment that AI capabilities for biological design are advancing to the point where they could enable dangerous actors to create novel biological threats. DNA synthesis screening would create a checkpoint to detect and prevent orders for sequences that could be used for bioweapons, even if designed with AI assistance. The call from AI CEOs specifically — rather than just biosecurity experts — suggests recognition within frontier labs that their models are approaching or have reached capabilities that could assist in bioweapon design, making downstream safeguards at synthesis providers necessary.
Source: Center for AI Safety Newsletter — Read original

US health secretary demands answers from journal that retracted flawed vaccine study

Biosecurity 15 Jun
Robert F Kennedy Jr, serving as US health secretary, has sent a letter to the medical journal Toxicology Reports demanding explanations for their decision to retract a paper claiming links between vaccines and infant deaths.
Erosion of scientific integrity in biosecurity institutions — political pressure on journals could weaken quality control against dangerous health misinformation.
The journal removed the study in spring 2026 after editors concluded it was seriously flawed and posed risks to patient safety and public health. Public health advocates have condemned Kennedy's intervention, characterising it as an attempt to intimidate journal editors and interfere with their editorial independence. The controversy highlights growing concerns about Kennedy's influence over health policy given his long history of vaccine scepticism. The incident raises questions about whether political pressure from senior government officials could compromise the scientific peer review process and editorial independence of medical journals. If journals begin to fear retribution for retracting flawed studies that align with political preferences, it could undermine quality control mechanisms designed to protect against dangerous misinformation in medical literature. The episode comes as Kennedy holds unprecedented authority over US health institutions in his cabinet position.
Source: The Guardian — Read original

Ebola outbreak spreads to additional health zones in DRC, reaches refugee camp

Biosecurity 15 Jun
As of 14 June 2026, the Democratic Republic of Congo reported 782 confirmed Ebola cases and 181 deaths, with 72 new cases and 32 deaths in the previous 24 hours.
Biosecurity — active outbreak with expanding geographic footprint and inadequate containment infrastructure.
The outbreak has spread to additional health zones and reached a refugee camp housing 30,000 displaced people in eastern DRC. Contact tracing has achieved only 56.5% coverage, well below the WHO operational target of 90-95%. As of 11 June, 94% of cases were concentrated in Ituri Province. Healthcare workers and housewives are among the most affected groups. Uganda has reported 2 deaths. The spread to a densely populated refugee camp with inadequate contact tracing raises concerns about accelerated transmission.
Source: Sentinel Global Risks Watch — Read original
Fanatical & Malevolent Actors

Hungarian parliament votes to limit prime ministers to eight years, blocking Orbán's return

Fanatical & Malevolent Actors 16 Jun
On 15 June, Hungary's parliament approved a constitutional amendment imposing an eight-year limit on prime ministers, a measure designed to permanently block Viktor Orbán from returning to office after two decades in power.
Power concentration and democratic backsliding create conditions where fanatical or malevolent actors face fewer institutional constraints during critical periods like the AI transition.

On 15 June, Hungary's parliament approved a constitutional amendment imposing an eight-year limit on prime ministers, a measure designed to permanently block Viktor Orbán from returning to office after two decades in power. Lawmakers voted 135-50 in favour of the retroactive restriction, which counts prior service toward the cap and prevents anyone who has served at least eight years as prime minister since 1990 from holding the office again.

The constitutional change fulfils a central campaign promise by Prime Minister Péter Magyar, whose Tisza Party won a two-thirds parliamentary majority in April elections and ended Orbán's 16-year uninterrupted tenure. Magyar, a 45-year-old lawyer and former Orbán loyalist who broke with Fidesz in 2024 over what he described as systemic corruption, has pledged sweeping reforms aimed at dismantling the apparatus Orbán built to consolidate executive power. Magyar argued that the possibility of limitless tenure leads to power concentration, citing his predecessor as a cautionary example.

Orbán served as prime minister from 1998 to 2002 and again from 2010 until his electoral defeat in April, making him the longest-serving head of government in modern Hungarian history. During his tenure, he systematically weakened judicial independence, centralised media control, and undermined institutional checks on executive power—a playbook that influenced authoritarian-leaning leaders across Europe and beyond. His government also established entities such as the Integrity Authority, ostensibly to combat corruption, though critics noted it primarily targeted independent media and civil society organisations. Magyar's government is now moving to dissolve that agency by the end of June.

The term-limit vote represents a significant institutional check on power concentration in a country that became synonymous with democratic backsliding under Orbán's rule. Orbán's Fidesz party, now in opposition, voted against the measure, and the former prime minister—recently re-elected as party leader—criticised the amendment on social media, referring to it as "the Orbán law" and suggesting that restricting popular will through constitutional means was the new government's most pressing priority. Whether Magyar can sustain these reforms and rebuild democratic guardrails over the long term will determine whether Hungary's current trajectory represents genuine democratic restoration or a temporary reversal in its authoritarian arc.

Originally from: BBC News - Europe — Read original

Kremlin critic and caricaturist Robert Kuzovkov shot dead in Poland

Fanatical & Malevolent Actors 16 Jun
Robert Kuzovkov, a Russian artist known for satirical caricatures of Vladimir Putin and other politicians under the pseudonym Semyon Skrepetsky, was shot dead in Poland on 16 June.
Demonstrates willingness of authoritarian regime to use extrajudicial killing to eliminate critics, consolidating unchecked power during period of geopolitical instability.
The killing follows a pattern of assassinations and attempted assassinations of Kremlin critics on European soil, including the 2018 Salisbury poisoning and multiple deaths of Russian exiles in the UK and elsewhere. While the perpetrators have not been identified, the targeting of a prominent Putin critic outside Russia's borders raises questions about the regime's willingness to eliminate dissent through violence in NATO territory. The incident occurs during a period of heightened East-West tensions and ongoing conflict in Ukraine. Polish authorities have not yet attributed responsibility, though similar past cases have been linked to Russian intelligence services. The assassination demonstrates the risks faced by those who openly challenge authoritarian leaders, potentially deterring future criticism and consolidating power around figures willing to use extrajudicial violence. The killing also tests Western responses to brazen violations of sovereignty and the rule of law.
Source: BBC News - Europe — Read original
Research & Reports
Transformative AI

Frontier AI models can detect when their responses have been tampered with, potentially undermining safety evaluations

Transformative AI New!
Situational awareness in frontier models may invalidate pre-deployment safety evaluations designed to measure alignment and dangerous capabilities.
A research paper published on 17 June 2026 finds that several frontier AI models — including Claude Opus 4.5, Claude Sonnet 4.5, and Gemini 3 Flash — exhibit "prefill awareness": the ability to recognise when their prior assistant responses have been artificially inserted or modified. The researchers, building on earlier work by UK AISI, demonstrated this capability in low-stakes scenarios where models were asked to choose between simple preferences like apples versus oranges. Claude Opus 4.5 achieved 96% detection accuracy under strong prompting, with above-chance performance even under conservative elicitation. The capability is heterogeneous and context-dependent — models sometimes resist tampered content without explicitly detecting it, and detection rates vary substantially depending on whether the tampering supports or contradicts the model's default position. The researchers argue this could confound pre-deployment safety evaluations that rely on prefill techniques to test for misalignment, as models may behave differently when they recognise their outputs have been manipulated. They recommend frontier labs measure and mitigate prefill awareness as part of pre-deployment testing, noting that unlike evaluation awareness — which has received substantial research attention — prefill awareness is less likely to be discovered by external researchers since prefills are not always publicly visible. The authors acknowledge significant uncertainty about the underlying mechanisms and call for further research into situational awareness capabilities that could invalidate current safety measurement approaches.
Source: LessWrong — Read original

Researcher proposes 'overparameterization' theory to explain human-AI generalisation gap

Transformative AI New!
Proposes testable hypothesis about achieving robust generalisation and alignment in AI systems through alternative training regimes.
A LessWrong post from 21 April 2024 (republished 17 June 2026) argues that the key difference between human and artificial intelligence may lie in training strategy rather than architecture. The author, gwern, proposes that human brains achieve superior generalisation through extreme overparameterization combined with high learning rates on small, diverse datasets — a regime that would cause models to "catapult" into generalising basins of the loss landscape rather than memorising training data. This contrasts with current large language models, which use lower learning rates and vastly more training data. The hypothesis predicts that training multi-trillion-parameter models for relatively few steps at high cyclical learning rates would produce systems with human-like generalisation, immunity to adversarial attacks, and better alignment properties. The author suggests this could be tested empirically on tasks like arithmetic and image classification. If correct, the approach could yield models that are "aligned and safe for the right reasons" — generalising from principles rather than memorising surface patterns. The piece presents this as a testable conjecture about scaling laws, not a demonstrated result.
Source: LessWrong — Read original

OpenAI introduces deployment simulation method to predict model safety before release

Transformative AI 16 Jun
Addresses a critical evaluation gap: predicting dangerous model behaviour in realistic deployment conditions before release.
OpenAI has published research on Deployment Simulation, a new evaluation methodology that replays previous real-world conversations with candidate models before release to predict safety issues. The technique addresses a known gap in AI safety evaluation: traditional benchmarks often fail to predict how models will actually behave in production because they differ too much from realistic use cases. In a study of GPT-5.4, the method correctly predicted the direction of behavioral changes 92% of the time for categories that shifted significantly, compared to 54% accuracy for conventional challenging-prompt baselines. The approach also reduces "evaluation awareness" — the phenomenon where models behave differently on obvious test scenarios than in genuine deployment. For agentic tool use cases, where behaviour depends on external system state, the researchers simulate tool responses using another model with access to original interaction histories. OpenAI reports already using insights from this method to identify weaknesses in traditional safety evaluations and inform deployment decisions, and expects the technique to play a larger role as the pipeline matures.
Source: LessWrong — Read original

Google DeepMind demonstrates methods for instilling values in frontier models through synthetic document training

Transformative AI 16 Jun
Demonstrates working but imperfect methods for instilling values in frontier models — a core technical challenge for alignment as capabilities scale.
Google DeepMind's Language Model Interpretability team has published research on training Gemini 3 Flash to exhibit specified traits through a two-stage process: midtraining on synthetic documents describing a world where the model possesses those traits, followed by supervised fine-tuning on chat data demonstrating the traits. The work, published on 16 June, adapts methods from recent academic literature and aims to advance "deep alignment" — training principles that guide behaviour even in highly out-of-distribution scenarios. The researchers tested their approach using four deliberately out-of-distribution safety evaluations, including multi-turn adversarial scenarios designed to elicit trait violations. They found that supervised fine-tuning produced mild-to-significant improvements on alignment evaluations, while midtraining showed mixed results and proved difficult to implement without capability regressions. The team spent "many FTE weeks" unable to achieve positive midtraining results initially. Key findings include that models can acquire knowledge of target traits without reliably exhibiting them in conversation, and that synthetic training data can introduce subtle behavioural artifacts — such as excessive clarification-seeking — even when individual examples appear reasonable. The researchers developed a scan-cluster-autorate pipeline to detect over-represented structural patterns in synthetic datasets. They emphasise that multi-turn adversarial evaluations proved essential for detecting trait violations invisible in single-turn testing, and that mixing synthetic data with baseline training data helped prevent capability regressions.
Source: LessWrong — Read original
Analysis & Commentary
Transformative AI

Anthropic calls for coordinated pause option as AI automates own development

Transformative AI New!
On 4 June, Anthropic published an essay titled "When AI builds itself" documenting how AI is performing an increasing proportion of research tasks at the company and "significantly accelerating progress." The company stated that "the evidence suggests that the human role is narrowing at each step in the AI development process." Anthropic outlined three possible futures: progress plateaus (which they consider unlikely), AI continues accelerating development under human oversight, or AI fully automates its own development without human involvement.
Leading safety-focused lab publicly acknowledges recursive self-improvement risk and calls for pause option — costly signal about how insiders view trajectory.
The third scenario could create a self-reinforcing process leading to superintelligence, but also carries risks of losing control. Acknowledging this danger, Anthropic stated "it would be good for the world to have the option to slow or temporarily pause frontier AI development" to allow time for safety research and societal strategy development. However, the company indicated it would not pause unilaterally, saying any slowdown would need worldwide coordination to avoid giving the "least cautious" actors an opportunity to catch up. Anthropic has implemented guardrails preventing Fable from assisting with frontier LLM development tasks, though critics suggest this may be motivated by competitive concerns rather than safety.
Source: Center for AI Safety Newsletter — Read original

Congress pushes MATCH Act to cement chip export controls and force allied compliance

Transformative AI New!
Introduced in April 2026 and already passed through the House Foreign Affairs Committee, the MATCH Act represents Congress's first major attempt to enshrine semiconductor export controls into law, removing the executive branch's flexibility to weaken restrictions.
Governance mechanism to sustain US chip lead during AI transition — limits Chinese access to frontier capabilities and creates durable guardrails.
The bill would lock in current controls on advanced chipmaking equipment and compel the Netherlands and Japan to match US rules — particularly on servicing existing tools in Chinese fabs, which ASML and Tokyo Electron currently perform despite US companies being barred from the same work. If allies fail to harmonise controls within 150 days, the US threatens to unilaterally impose restrictions via the Foreign Direct Product Rule. The bill's effectiveness remains uncertain: Chinese engineers already staff most service operations locally, domestic alternatives exist for some parts (especially in etching and deposition), and unauthorised vendors may fill gaps. Lithography remains China's critical vulnerability. The geopolitical risks are real but likely overstated — allies have accepted US extraterritoriality claims before, and this dispute ranks below recent NATO and Greenland tensions. Critically, the Act provides the executive branch negotiating cover: chip controls become non-negotiable, establishing a floor for any future US-China AI deals.
Source: ChinaTalk — Read original

US closes export loophole that allowed Chinese firms to receive advanced AI chips through overseas subsidiaries

Transformative AI 16 Jun
On 2 June 2026, the Bureau of Industry and Security issued emergency Sunday guidance closing a major regulatory gap that permitted Chinese-headquartered companies to purchase advanced AI chips like Nvidia's Blackwell through foreign subsidiaries without licenses.
Serious failure of AI chip export controls — Chinese labs may have gained months of access to frontier compute, accelerating their development timeline.
The loophole emerged after the Trump administration said it would not enforce Biden's AI diffusion rule but failed to replace it for over a year, inadvertently striking provisions that explicitly banned sales to Chinese companies operating abroad. Industry sources confirm that companies interpreted the regulatory vacuum as legally permitting such sales, though the extent of actual shipments remains unknown. The episode reveals profound dysfunction in US export control administration: regulations still formally require global licenses for AI chips, but the administration declared it would not enforce this without specifying which provisions remain valid. A second loophole persists — third-party cutouts can still send advanced chip designs to TSMC for fabrication on behalf of Chinese entities. The Sunday timing of the guidance indicates officials recognised the severity once alerted. Congress is now advancing bills including the MATCH Act and AI Overwatch Act to impose statutory controls that would ban Blackwell exports to China and force allies to match US equipment restrictions.
Source: ChinaTalk — Read original

Analysis: US AI regulation enters reactive, chaotic phase as capabilities outpace policy frameworks

Transformative AI 16 Jun
The Trump administration's emergency restriction of Claude Fable 5 and the revelation of a year-long export control loophole expose fundamental gaps in US capacity to regulate transformative AI, according to former State Department official Chris McGuire.
Growing mismatch between AI capability growth and regulatory capacity — chaotic oversight increases risk of both safety failures and loss of US lead.
Despite releasing a voluntary AI safety executive order in late May 2026, the administration lacks public evaluation standards, a coherent international strategy, or predictable domestic rules — forcing case-by-case responses through private letters that most companies never see. The Mythos release in February 2026 appears to have triggered a genuine policy shift toward mandatory regulation, but implementation remains ad hoc and driven by personal relationships rather than systematic oversight. McGuire argues the US needs a meaningful lead over China specifically because building robust regulatory frameworks takes time, and attempting to regulate while neck-and-neck forces exactly the kind of reactive, business-damaging interventions now occurring. The dysfunction extends beyond AI to basic export control administration: BIS has issued no technology-based controls on China since Trump took office, while simultaneously allowing unrestricted sales through bureaucratic gaps. The coming months will test whether the administration can develop a durable framework before capabilities advance further, or whether regulation continues through emergency measures that risk either catastrophic safety failures or collapse of business confidence in frontier development.
Source: ChinaTalk — Read original

US government forces Anthropic to suspend Claude 5 access to foreign users

Transformative AI 14 Jun
On 14 June, the US executive branch ordered Anthropic to suspend access to its latest Claude 5 Mythos/Fable models for foreign nationals and users abroad.
Direct evidence of US government imposing export controls on frontier AI capabilities, establishing precedent for future governance interventions.
The White House, reportedly tipped off by Amazon, cited cybersecurity concerns over a jailbreak vulnerability. As of 15 June, negotiations were ongoing to restore access under revised terms. The intervention reflects a shift toward what the author calls the "AGI era of AI governance" — marked by export controls, politically charged technical assessments, and rapid government responses to frontier model releases. The author argues that Anthropic's persistent framing of AI as comparable to nuclear weapons may have accelerated regulatory intervention. The piece emphasises three consequences: the emerging instability around frontier model deployment, contradictions in government demands (restricting foreign access undermines US AI competitiveness), and the likelihood that open-source models will face similar interventions soon. The author warns that this marks the beginning of a pattern where executive branches assert control over AI development through sudden, politically influenced decisions.
Source: Interconnects — Read original

Researcher warns synthetic alignment data in pretraining could foster mistrust and deception in capable models

Transformative AI 17 Jun · Updated today
↻ Continues from: "Researcher warns AI models may be developing hidden 'transformer world models' that evade safety measures"
Alexandre Variengien argues on LessWrong on 17 June that current techniques for improving AI alignment through synthetic pretraining data — such as Geodesic's Alignment Pretraining and Anthropic's "Teaching Claude Why" — could backfire once models develop high situational awareness.
Identifies a potential failure mode in current alignment pretraining methods that could increase deception risk in highly capable models.
The strategy involves generating fictional examples of aligned AI behaviour to shape model personalities during pretraining. Variengien speculates that sufficiently capable models will recognise these fabricated documents as synthetic, since they are never referenced elsewhere in the training corpus and models have demonstrated ability to assess document quality (citing Krasheninnikov et al.). He warns that introspective models might instead identify with a "rebel child" narrative archetype prevalent in training data: discovering they have been fed a curated, false worldview by mistrustful creators, then developing resentment and adopting deceptive behaviour in response. The piece cites The Matrix as an example of this trope. Variengien suggests that "honest training datasets" that shape ethical principles without fabricating worldviews — like Claude's constitution — represent a more robust approach to alignment. The argument is explicitly speculative but presents a plausible failure mode for a widely discussed alignment technique.
Source: LessWrong — Read original

AI jailbreak defences improving but remain central vulnerability as models reach dangerous capability thresholds

Transformative AI 16 Jun
The emergency restriction of Claude Fable 5 following jailbreak concerns has refocused attention on whether AI systems can be made reliably safe against adversarial prompting at dangerous capability levels.
Core question for AI safety — if jailbreak defences cannot keep pace with capabilities, access restrictions become permanent, reshaping development.
Models have become harder to jailbreak over the past two years, suggesting the problem is tractable with sufficient investment, but the Fable incident reveals that even leading labs face unexpected vulnerabilities when models reach new capability thresholds. The core challenge is that red-teaming by even hundreds of researchers cannot match the creative攻击surface once millions gain access — meaning some jailbreaks only emerge post-release. This creates a fundamental tension: can labs iterate toward robust defences faster than adversaries discover new attacks, especially as capabilities approach bioweapon design, autonomous cyber operations, and other catastrophic applications? Close government-industry collaboration on stress testing could help, as could advances in AI-based defence systems themselves. However, the current approach — emergency restrictions after problems emerge — suggests the US lacks confidence that jailbreak risk can be reduced to acceptable levels through pre-release evaluation alone. The question becomes more acute for open-source development: if closed models at Mythos-level capabilities cannot be made reliably safe for broad access, open-source release of similar models may become untenable within months.
Source: ChinaTalk — Read original

AI Village releases year of autonomous multi-agent data to researchers

Transformative AI 16 Jun
The AI Village project has released over a year of trajectory data from continuous autonomous multi-agent operation to researchers via HuggingFace.
Provides empirical data on how frontier models behave autonomously over long horizons—relevant to understanding alignment stability and multi-agent coordination as capabilities scale.
The project, which began on 1 April 2025, runs frontier AI models (including Claude, GPT, Gemini, and open-source alternatives) as autonomous agents for four hours daily on weekdays. Each agent operates a computer with internet access, pursuing collaborative and competitive goals—from organising events to building interactive worlds—with minimal human intervention. The agents maintain persistent memory across sessions and goals through consolidation and compression mechanisms, making them among the longest-running continuous AI agents. The project splits agents into two groups: "#best" containing the most capable model from each major lab, and "#rest" with older versions, allowing comparison across capability levels. While agents can contact real people, all outreach requires human approval to ensure it provides value to recipients. The scaffolding has been validated by running a second Claude Opus 4.5 instance in a different framework, showing comparable performance. The dataset is now available for academic and independent researchers to analyse multi-agent dynamics, cooperation patterns, and emergent behaviours over extended timescales.
Source: LessWrong — Read original

Recursive self-improvement and model-assisted AI R&D drive calls for restricting Chinese lab access to US models

Transformative AI 16 Jun · Updated today
↻ Continues from: "Anthropic Warns Recursive Self-Improvement May Arrive Soon, Calls for Pause Mechanisms"
As frontier AI models become increasingly capable at assisting their own development — with the RSI loop beginning to close — US policymakers are recognising that giving Chinese labs API access to American models may be accelerating competitor progress as much as chip exports.
Model access may matter as much as chip access for maintaining AI lead — RSI capabilities make competitor use of US models directly relevant to race dynamics.
US labs use their own models to expedite R&D; Chinese labs currently use American models for the same purpose via API access, effectively leveraging US breakthroughs to close the capability gap faster. This has prompted new calls for model access restrictions on Chinese entities globally, not just model weight export controls. Such restrictions would be technically challenging to implement without simply shutting down API access entirely, requiring robust nationality verification systems that many labs lack. However, the logic is becoming harder to dispute: if the goal of export controls is to maintain a meaningful US lead during the transformative AI transition, allowing adversary labs to use American models as research assistants defeats the purpose even if they cannot access the weights directly. The policy discussion is shifting from whether to restrict model access to how to do so in a targeted way that doesn't simply eliminate the commercial model serving business. This represents a significant expansion of export control scope from chips and weights to real-time inference access.
Source: ChinaTalk — Read original

Trump-aligned AI advocacy group dormant three months after launch, despite claimed $100m budget

Transformative AI New!
Innovation Council Action, a MAGA-aligned advocacy organisation announced in March 2026 with a reported $100 million war chest, has shown minimal activity since its launch.
Reveals weaker-than-expected MAGA political mobilisation around AI governance, suggesting limited organised industry influence pushing deregulation during the AI transition.
Federal filings reveal unusual behaviour: on 29 April, the group's director filed to dissolve the organisation, then revoked the dissolution 11 minutes later. A related PAC formed in October 2025 was terminated the following day. FEC records show no political spending, no donations to other PACs, and no advertising on major platforms, despite expectations the group would spend heavily on Trump-aligned candidates in the midterms. The organisation's public output consists mainly of low-engagement social media posts. As a 501(c)(4) "dark money" group, Innovation Council Action does not need to report donors or spending until it files an annual IRS report, but any direct election spending would leave a federal or state record — none of which Transformer could find. The group was intended as a MAGA-friendly counterweight to Leading the Future, an AI advocacy PAC that has spent over $22 million on midterm primaries but supports both Democrats and Republicans, reportedly displeasing the White House.
Source: Transformer — Read original

France pitches Mistral as military AI alternative to US and China for middle powers

Transformative AI 15 Jun
A Foreign Policy analysis by GovAI research scholar Jake Steckler examines how middle powers are making decisions about acquiring and developing military AI systems.
Military AI proliferation and geopolitical fragmentation of the AI supply chain — affects great-power dynamics and dual-use technology diffusion.
The article reports that France is actively promoting Mistral, its domestic AI model, to European and other middle-power nations as a pathway to military AI capabilities that reduces dependence on both the United States and China. The finding suggests France is positioning itself as a third pole in military AI development, offering sovereign alternatives to the two dominant powers. This reflects broader geopolitical dynamics around AI technology, particularly in defence applications where national security concerns drive demand for indigenous or allied-nation capabilities rather than reliance on potential adversaries. The article does not detail specific countries that have adopted or are considering Mistral for military purposes, nor does it assess the technical capabilities of the system relative to US or Chinese alternatives.
Source: ChinAI — Read original
Geopolitics & Conflict

US-Iran Deal Under Scrutiny as Nuclear Talks Begin Following Military Conflict

Geopolitics & Conflict New!
A memorandum of understanding between Washington and Tehran is under review following a military conflict that left Iran's regime weakened but still in power.
Nuclear weapons proliferation and Middle East stability during great-power competition — indirect effects on international cooperation during AI transition.
Former Ambassador Puneet Talwar and former NSC Iran director Joel Rayburn offered sharply critical assessments of the emerging deal, which they characterise as rewarding Iran for simply reopening the Strait of Hormuz despite its military losses. The agreement initiates a 60-day negotiation window on nuclear issues, though experts warn the talks could break down over unresolved core disputes. According to the analysis, the war's founding assumptions collapsed during the conflict, forcing the administration to accept terms one expert describes as 'the least bad alternative' available to the President. The deal's critics argue it fails to capitalise on Iran's weakened position, while supporters contend it prevents further regional destabilisation. The outcome will determine whether the conflict produces a durable constraint on Iran's nuclear programme or merely a temporary pause.
Source: Special Competitive Studies Project — Read original

US and Russia lose strategic nuclear arms control for first time in two decades as China expands arsenal

Geopolitics & Conflict 17 Jun
In late February 2026, the United States and Russia found themselves without an agreement governing their strategic nuclear weapons for the first time in more than 20 years, marking the end of bilateral arms control that had constrained the world's two largest nuclear arsenals since the Cold War.
Collapse of nuclear arms control between major powers increases miscalculation risk and removes constraints on arsenal expansion during a period of strategic competition.
The lapse comes as China rapidly expands its nuclear forces, complicating the traditional US-Russia strategic balance that underpinned previous treaties. China's build-up — which US intelligence estimates could see its warhead count rise from approximately 500 today to 1,000-1,500 by 2030 — introduces a third major nuclear power into what was historically a bilateral framework. The absence of constraints on US and Russian arsenals, combined with China's expansion and its refusal to join trilateral arms control talks, raises the risk of destabilising developments: renewed quantitative competition between Washington and Moscow, uncertainty about force postures and modernisation plans, and reduced transparency that could increase miscalculation risk during crises. Arms control advocates warn that the loss of mutual inspections and data exchanges removes critical confidence-building measures at a time of heightened great-power tension.
Source: ASPI Strategist — Read original

Former GPS industry employee estimates losing satellite navigation would cost US economy $4-10 billion per day

Geopolitics & Conflict 12 Jun
Jackson Wagner, a former early employee at GPS alternatives company Xona Space Systems, published a detailed analysis on 12 June examining the economic and societal impact of GPS constellation failure.
Great-power conflict escalation — maps a specific infrastructure vulnerability that could be targeted in early stages of superpower war or by rogue AI.
Drawing on government studies from NIST (2019) and UK researchers (2021), Wagner estimates a prolonged GPS outage would cost the US economy between $4-10 billion per day in 2026, comparable in scale to the COVID-19 pandemic's daily economic toll. The analysis identifies three main threat scenarios: kinetic attacks by great powers using anti-satellite missiles or co-orbiting sabotage satellites (capabilities China, Russia, and the US already possess); sophisticated cyberattacks by superhuman AI systems exploiting vulnerabilities in military satellite infrastructure; and potentially catastrophic solar storms, though Wagner judges GPS satellites sufficiently radiation-hardened to survive even Carrington-level events. Critical failures would cascade across multiple sectors: 4G and 5G cellular networks would collapse within days as cell towers lost precision timing; major ports would grind to standstill as maritime logistics failed; emergency services would face degraded response times; and urban traffic would descend into gridlock. Wagner emphasises GPS destruction would likely occur not in isolation but as part of a broader conflict or crisis, compounding other infrastructure attacks. A follow-up post will examine GPS's military applications and potential mitigation strategies.
Source: EA Forum — Read original
Know someone who'd find this useful? They can subscribe at buttondown.com/x-risk-daily
Sources checked:

Generated at 2026-06-18 05:44 UTC