The majority of code at Anthropic is now written by Claude, with some employees ceasing to write code directly and instead focusing on oversight and validation of AI-generated output.
Clark, who heads the Anthropic Institute launched in March 2026, said he returned from paternity leave in February to discover colleagues had fundamentally reoriented their roles around managing AI systems and verifying their outputs rather than producing code themselves. The internal shift has led to an explosion in code volume, prompting greater investment in telemetry and observability infrastructure to monitor what Clark characterised as an emergent ecosystem of autonomous agents. In one experimental deployment focused on automated alignment research, a single human researcher effectively supervised a team of nine synthetic research agents working on AI safety problems.
The transformation has reshaped Anthropic's hiring priorities. Clark noted that the value of junior engineering talent is becoming "a bit more dubious" inside the company, while senior engineers with "really, really well-calibrated intuitions and taste" are increasingly valuable. The company now seeks early-career candidates with deep expertise in large language models and highly experienced professionals capable of conceptualising ambitious projects, rather than mid-level engineers focused on implementation. Clark predicted that AI could account for 99% of the company's coding by year-end "if things speed up really aggressively".
According to external research organisation METR, Claude Opus 4.6 can reliably complete tasks that would take a skilled human approximately 12 hours, representing a dramatic compression of capability timelines from earlier models. Anthropic reports the model "plans more carefully, sustains agentic tasks for longer, can operate more reliably in larger codebases, and has better code review and debugging skills". The system features a one-million-token context window and introduced agent team functionality, allowing multiple Claude instances to coordinate on complex engineering tasks.
Clark framed the internal experiment as Anthropic deliberately stress-testing its own systems ahead of more capable models. In an interview with Axios, he placed the development in a broader trajectory, predicting a 60% or greater probability that an AI model will autonomously train its successor by the end of 2028. He characterised the shift as humans moving to a verification layer atop a vastly expanded virtual organisation of AI systems, with teams expected to shrink in headcount while tackling more ambitious technical objectives. Anthropic's research agenda warned of "AI contributing to speeding up the research and development of AI itself", a dynamic known as recursive self-improvement that has historically been confined to theoretical AI safety literature.
On 26 May, the Office of Personnel Management posted a draft rule to the Federal Register proposing government-wide non-disclosure agreements for all federal employees, both new hires and existing staff. The move marks an unprecedented expansion of secrecy requirements across the roughly 2 million-strong federal workforce, where such agreements have historically been confined to classified or national security positions.
The proposed NDAs would cover information relating to internal agency operations, personnel matters, procurement processes, and any sensitive, pre-decisional or deliberative material not currently publicly available, according to CNN. The administration justified the measure by citing recent leaks about immigration enforcement operations and the secretive January raid on Venezuela that captured former President Nicolás Maduro, which officials claim endangered the lives of federal agents and military personnel. The draft agreement would allow the government to pursue civil and criminal penalties against employees who disclose covered information, and grants the administration rights to royalties received from such disclosures—a provision whose practical application remains unclear. Former employees would require written permission from authorized agency officials to speak publicly about confidential information even after leaving government service, with the agreement remaining effective for five years post-employment.
While the draft explicitly preserves rights to make disclosures authorized by law, including under the Whistleblower Protection Act, civil liberties advocates and federal unions have raised constitutional concerns. Everett Kelley, national president of the American Federation of Government Employees, characterized the proposal as an attempt to purge nonpartisan career employees and replace them with loyalists unwilling to report waste, fraud, and abuse. Ray Limon, who served as a federal government attorney and human resources leader for nearly three decades, told NPR that such broad language could discourage lawful whistleblower disclosures despite the stated protections. Lauren Harper of the Freedom of the Press Foundation described the policy as "dangerously secretive," warning it would undermine transparency mechanisms that have historically exposed government wrongdoing.
The proposal, which enters a 30-day public comment period following its Federal Register publication, would permit individual agencies to decide whether to implement the agreements. A separate draft rule proposed in 2025 suggested that refusal to sign could result in termination or debarment from future federal employment. OPM Director Scott Kupor defended the measure by comparing it to private-sector confidentiality agreements, stating that the federal government should not be held to a lower standard in protecting sensitive information. The initiative forms part of the Trump administration's broader campaign to control information flows from federal agencies, following earlier moves including NDAs imposed at the Pentagon in October 2025 and restrictions on press access to military facilities.
An Anthropic spokesperson told TechCrunch that Karpathy will start a team focused on using Claude to accelerate pre-training research, signaling an intensifying race among frontier labs to develop AI systems capable of improving their own capabilities.
Karpathy began work this week on Anthropic's pretraining team under team lead Nick Joseph, another OpenAI alumnus. Pretraining is responsible for the large-scale training runs that give Claude its core knowledge and capabilities, and is one of the most expensive, compute-intensive phases of building a frontier model. The move represents a significant talent acquisition in what Axios described as "a major coup for Anthropic in the escalating competition for elite AI talent".
Karpathy's appointment comes amid a broader pattern of senior technical leaders joining Anthropic in individual contributor research roles. CTOs of billion-dollar companies have been quitting to take individual contributor roles at Anthropic, including the CTOs of Workday, You.com, Instagram, Box, Super.com, and Adept AI between mid-2025 and early 2026. The concentration of talent has not gone unnoticed: Karpathy is one of the few researchers who can bridge the gap between LLM theory and large-scale training practice, and tapping him to build such a team is a clear sign from Anthropic that it believes AI-assisted research, rather than pure compute, is how it stays competitive with OpenAI and Google.
The focus on recursive self-improvement has sparked controversy within the AI safety community, with researcher Nate Soares calling it "not 'good guys' behavior" to hire top scientists to work on potentially dangerous technology. The concerns center on systems that could amplify their own capabilities without human oversight. Anthropic co-founder Jack Clark had predicted in early May a 60% chance of full recursive self-improvement by the end of 2028, according to The Algorithmic Bridge. Industry reactions ranged from sports analogies comparing the hire to superstar free agency moves to deeper concerns about the wisdom of accelerating work on self-improving AI systems.
Karpathy had previously left OpenAI and worked on AI education initiatives, including founding Eureka Labs and creating the widely-followed "Neural Networks: Zero to Hero" educational series. He stated he remains "deeply passionate about education and plan[s] to resume [his] work on it in time". Anthropic has been in discussions on a $30 billion fundraising round that would value the company at $900 billion, surpassing rival OpenAI's most recent valuation of $852 billion, according to reports from multiple outlets tracking the AI funding landscape.
Speaking at Bletchley Park, the World War II code-breaking centre 45 miles northwest of London, Keast-Butler characterised the current environment as a "new era of radical uncertainty, contested geopolitics and rapidly changing technology", and warned that "the risk of miscalculation is as high as I've ever seen it".
The GCHQ director detailed the scope of Russian operations against the UK and Europe, stating that Moscow is "relentlessly targeting critical infrastructure, democratic processes, supply chains and public trust". Her remarks come as Western intelligence agencies increasingly warn that Russia is stepping up hostile activity in a "gray zone" that falls just below the threshold of war, with authorities in Sweden, Poland, Denmark and Norway alleging that Russian-linked hackers targeted critical infrastructure including power plants and dams. Keast-Butler also accused Russia of stealing technology and plotting sabotage and assassination attempts, highlighting GCHQ's work in disrupting Russia's efforts to smuggle Western technology.
On China, the intelligence chief warned that rapid advances in artificial intelligence mean "the ground beneath our feet is shifting" and there is a "narrowing window for the U.K. and allies to stay ahead" of countries such as China, described as a science and technology "superpower". She called for a dramatic shift in how both private sector and citizens approach cybersecurity, arguing there must be an effort "from boardrooms to living rooms" to make cybersecurity "10 times more urgent". This echoes recent warnings from the head of the National Cyber Security Centre, which is part of GCHQ, who cautioned last month that Britain should brace for a rise in cyberattacks linked to hostile states.
The choice of venue carried symbolic weight: Bletchley Park, where hundreds of mathematicians and cryptographers worked to crack Nazi Germany's secret codes in work that both shortened the war and hastened the birth of modern computing. The speech also marked the 80th anniversary of the UK-US intelligence agreement, though Keast-Butler's emphasis on international partnerships comes as U.S. President Donald Trump's "America First" foreign policy strains the relationship between London and Washington.
The framing of both threats together — Russia's active sabotage efforts and China's accelerating technological development — underscores British intelligence's assessment that the strategic environment has fundamentally deteriorated, with the risks of miscalculation between major powers now at levels unprecedented in Keast-Butler's career overseeing electronic intelligence operations.
On 20 May, OpenAI announced that an internal reasoning model had autonomously disproved the planar unit distance conjecture, an 80-year-old problem in discrete geometry first posed by Paul Erdős in 1946. The problem asks: if you place n points in a plane, what is the maximum number of pairs that can be exactly distance 1 apart? For decades, mathematicians believed square grid arrangements were essentially optimal. The OpenAI model disproved this longstanding conjecture, discovering an infinite family of constructions using deep algebraic number theory (Golod-Shafarevich theory, infinite class field towers) that achieve polynomial improvement over square grids—specifically, n^(1+δ) unit-distance pairs for some fixed δ > 0 (later refined to δ = 0.014 by Princeton mathematician Will Sawin). The proof has been checked by a group of external mathematicians. They have also written a companion paper explaining the argument and providing further background and context for the significance of the result. A human-verified version of the proof has been published on arXiv.
The announcement marks a departure from OpenAI's October 2025 controversy, when the company claimed GPT-5 had cracked ten Erdős problems. Thomas Bloom, the same mathematician who has now verified the unit-distance result, called the framing a "dramatic misrepresentation" at the time. The model had retrieved solutions, not produced them. The May 2026 result has received far more positive reception. Tim Gowers, a Fields Medal winner, called the result "a milestone in AI mathematics." Nature and other outlets have noted strong reactions from the mathematics community, though OpenAI has not revealed all procedural details.
Separately, Google DeepMind released AlphaProof Nexus, a system combining large language models with the Lean formal proof assistant. The system autonomously solved 9 out of 353 open problems from the Erdős catalog — a collection of unsolved mathematical challenges — at an inference cost of just a few hundred dollars per problem. Two of the nine problems had been open for 56 years. The results were documented in an arXiv preprint (2605.22763v1) published on 21 May 2026. All formal proofs and selected natural language versions have been made available in a GitHub repository that was updated between 20 and 22 May 2026. The system also proved 44 of 492 open conjectures from the Online Encyclopedia of Integer Sequences.
The dual announcements represent a significant shift in AI's mathematical capabilities. Competition math is difficult, but it is designed to have a clean answer. Open research problems are different. They may have resisted specialists for decades, they may require a new construction, and they do not come with the comfort of knowing that a solution is waiting at the end of the page. For those tracking transformative AI development, the achievements signal that general-purpose models can now autonomously make progress on genuinely difficult, open-ended problems requiring creative insight—a capability with direct implications for recursive self-improvement pathways. However, the practical impact remains uncertain. The OpenAI solution procedure was described in the original announcement as complicated and probabilistic enough that it cannot be run in practice, raising questions about whether such breakthroughs are immediately useful or primarily symbolic demonstrations of capability.
Go deeper: Remarks on the disproof of the unit distance conjecture (arXiv, verified proof)
The signing ceremony, scheduled for Thursday afternoon with invited technology executives, was cancelled hours before it was set to begin after the president expressed concerns about impeding American competitiveness with China.
According to CNN, the draft order was divided into two sections: cybersecurity protections and "covered frontier models." The framework would have encouraged AI companies including OpenAI and Anthropic to provide early government access for security review before deploying advanced systems publicly. Trump told reporters he "didn't like certain aspects" of the order and worried it "could've been a blocker," adding that he did not want to do anything that might undermine the US lead over China in AI development.
The decision came after reported conversations with technology industry figures including former White House AI adviser David Sacks, who favours a hands-off regulatory approach, as well as Elon Musk and Mark Zuckerberg, according to The Hill. The reversal marks a significant victory for AI laboratories seeking to avoid government oversight and a setback for safety advocates who have pressed for pre-deployment scrutiny of high-risk systems. While the framework would not have been legally binding, it would have established a precedent for government review of frontier models — a mechanism advocated in many proposed governance frameworks for advanced AI.
The timing of the postponement is notable given recent releases of increasingly capable AI models with cybersecurity implications. The Washington Post reported that officials had become concerned about new systems like Anthropic's Mythos, which have demonstrated capabilities in identifying and exploiting software vulnerabilities. The reversal exposes tensions within the administration between those prioritising economic and strategic competition with China and those concerned about the risks of deploying powerful systems without adequate evaluation, suggesting that competitiveness considerations currently outweigh safety concerns in White House AI policy.
The disclosure came as part of SpaceX's preparations for a June 12, 2026 listing targeting a valuation between $1.75 trillion and as high as $1.75 trillion, which would make it the largest IPO in history.
The Anthropic agreement, announced in early May but without initial financial details, grants the AI lab access to more than 300 megawatts of new capacity (over 220,000 Nvidia GPUs) across SpaceX's Colossus data centre facilities in Memphis, Tennessee. Anthropic announced moments before the filing became public that it was expanding beyond SpaceX's Colossus 1 facility to Colossus 2 as well. The deal allows either Anthropic or SpaceX to exit with 90 days' notice, and SpaceX indicated in the filing that it expects to enter into additional similar services contracts.
The arrangement illustrates what some in the industry call a "neocloud" model, which lets AI companies offset infrastructure costs by acting as a cloud provider when their own usage falls short of capacity. SpaceX's S-1 filing shows the company lost nearly $5 billion in 2025, with its AI division xAI — which merged with SpaceX in February 2026 — losing $6.4 billion. The company is spending $2.8 billion on gas turbines for its Colossus data centres and plans to scale its Grok model to multiple trillions of parameters while pursuing ambitions to launch data centres into space by 2028.
The filing also disclosed substantial financial entanglements within Elon Musk's corporate ecosystem, including a January 2026 arrangement in which Tesla agreed to invest $2 billion in xAI through a purchase of Series E Redeemable Convertible Preferred Stock, which was later converted to SpaceX equity following the merger. SpaceX cited AI backlash as a potential risk factor and set aside $530 million for potential litigation over features like Grok's "Spicy" and "Unhinged" modes. AI safety organisations published a letter warning that xAI's poor safety record could complicate fundraising. For Anthropic, the deal addresses acute capacity constraints that had led to aggressive rate caps for developers, with the company stating the additional compute would directly improve capacity for Claude Pro and Claude Max subscribers.
The company is working with Goldman Sachs and Morgan Stanley to prepare the draft prospectus, with a public debut targeted for as early as September 2026.
The move comes despite staggering financial losses. OpenAI generated $5.7 billion in revenue during the first quarter of 2026 but reported an adjusted operating margin of negative 122 percent, meaning the company lost $1.22 for every dollar of revenue earned. CEO Sam Altman reportedly told staff this week that filing for an IPO is different from being ready to go public, and that the company would not list until prepared. The company was last valued at more than $850 billion by private investors, though analysts expect it could be valued at up to $1 trillion by the time it goes public.
The listing arrives as OpenAI faces mounting pressure to demonstrate financial sustainability while investing heavily in AI infrastructure. The company has raised more than $180 billion from investors and continues to burn through cash at a historic pace. The company recently launched Guaranteed Capacity, which secures customers' compute access through one-to-three-year commitments, and announced a partnership with Malta to provide free ChatGPT Plus to all citizens completing a government AI literacy course — the first such national agreement. OpenAI also offered $2 million in tokens to every startup in the current Y Combinator batch in exchange for equity, signaling an aggressive push for market presence.
Altman is under pressure from investors to show that the numbers work while facing increasingly stiff competition from rivals, most notably Anthropic, which is currently in talks with investors to raise money at a $900 billion valuation. The IPO plan comes two days after OpenAI fended off an existential court challenge from Elon Musk, whose SpaceX filed confidentially for its own IPO in April and is expected to publicly disclose its prospectus shortly.
An IPO would subject OpenAI to unprecedented scrutiny and quarterly earnings pressures that could conflict with its stated long-term safety commitments. The company will likely have to address standard IPO questions such as competition and capital requirements, but OpenAI's own executives have repeatedly acknowledged that their technology might help people construct bioweapons and orchestrate massive coordinated cyberattacks. Companies filing confidentially receive feedback from the SEC before making their S-1 public, but the document must be published at least 15 days before the company begins its roadshow to sell shares to investors.
Munir's visit came amid reports that a peace deal between the United States and Iran had been almost finalised, with the army chief coordinating closely with Pakistan's Interior Minister Mohsin Naqvi, who had been in Tehran since 21 May holding detailed talks with Iranian leadership.
The engagement represents a significant regional mediation effort in a war that began on 28 February 2026 when the United States and Israel launched airstrikes on Iran, targeting military and government sites. After more than five weeks of fighting, the United States and Iran agreed on 7-8 April to a ceasefire that included Israel, but six weeks since the fragile ceasefire took effect, talks to end the war have made little progress. Pakistan has played a mediating role since April, with Prime Minister Shehbaz Sharif tasking Munir with maintaining behind-the-scenes contacts with American and Iranian political and military leadership, including all-night communications with US Vice President JD Vance, US special envoy Steve Witkoff and Iranian Foreign Minister Abbas Araqchi.
The conflict carries serious escalation risks given Iran's nuclear programme and the involvement of major powers. The surprise attacks launched during negotiations between Iran and the US assassinated several Iranian officials, including Supreme Leader Ali Khamenei. Iran responded with missile and drone strikes on Israel, US bases, and US-allied Arab countries, and closed the Strait of Hormuz, disrupting global trade. According to Al Jazeera, the conflict has resulted in thousands of casualties across the region, with Iran's Ministry of Health reporting at least 3,468 killed in US-Israeli attacks on Iran since February.
Pakistan's military leadership taking direct diplomatic action, rather than routing efforts through civilian foreign ministry channels, underscores the gravity of the situation. Field Marshal Munir held intensive talks with Iran's Parliament Speaker Baqir Qalibaf as well as Iran's chief negotiator, aiming to finalise a memorandum that would conclude hostilities. According to Reuters, Pakistan stepped up diplomatic efforts as President Donald Trump suggested he could wait a few days for "the right answers" from Tehran but was also willing to resume attacks. Qatar's parallel involvement in mediation, alongside Pakistani efforts acknowledged by the UK Parliament, suggests coordinated regional diplomacy to prevent further escalation in a conflict that has already triggered severe disruption to global energy markets and raised concerns about nuclear proliferation.
The United States has agreed to unfreeze billions of dollars in Iranian assets as part of a framework peace agreement with Iran's increasingly hardline government, according to reports confirmed by the Financial Times on 23 May. The preliminary deal, announced by President Trump as "largely negotiated" following talks with regional leaders including Saudi Arabia, Pakistan, and Israel, is expected to be finalised in the coming days.
The agreement has drawn sharp criticism not only from Republican foreign policy hawks, who traditionally support assertive US engagement with Iran, but also from Democrats concerned about the terms. Senator Cory Booker, speaking on CNN, noted that Trump previously criticised President Obama's 2015 nuclear deal for releasing $50 billion to Iran, yet "the president's balance sheet is letting more than $14 billion go through." According to Axios, earlier negotiations discussed unfreezing up to $20 billion in Iranian funds held in foreign banks, though the exact figure remains under negotiation. Euronews reports that Iran holds over $100 billion in frozen assets globally, making their release a central Iranian demand.
The proposed deal, mediated by Pakistan and Qatar, would establish a 60-day ceasefire extension during which the Strait of Hormuz—closed since the conflict erupted on 28 February—would reopen, and Iran would be permitted to sell oil freely. According to Axios, this initial phase would set the stage for broader negotiations on Iran's nuclear programme. However, crucial details remain unresolved, particularly regarding Iran's highly enriched uranium stockpile. NPR reported that a senior Israeli official characterised the emerging agreement as "bad because it signals to the Iranians that they possess a weapon no less effective than a nuclear one, and that is the Strait of Hormuz."
The timing of the potential agreement coincides with Iran's annual Khorramshahr liberation commemorations on 24 May, marking the 1982 victory during the Iran-Iraq War. Some Iranians view the prospect of sanctions relief and unfrozen assets as a historic turning point, particularly given the country's severe economic crisis. Iran's inflation reached 68.1 per cent in February, according to Euronews, the highest since the Second World War. Critics warn, however, that the deal may lack adequate safeguards. Former Secretary of State Mike Pompeo argued on social media that the agreement would enable Iran to "build a WMD program and terrorize the world," while Texas Senator Ted Cruz expressed concern about providing billions to a regime "still run by Islamists who chant 'death to America.'"
The US Department of Justice has removed news releases documenting criminal prosecutions of January 6 Capitol rioters from its website, describing the records as partisan propaganda. A review by NBC News found that the vast majority of press releases pertaining to Jan. 6 defendants have been removed from the DOJ website, eliminating official documentation of charges, convictions, and sentencings related to the 2021 attack, when Trump supporters stormed the Capitol attempting to prevent congressional certification of Biden's electoral victory.
The deletion came to public attention on 23 May when Washington Post reporter Meryl Kornfield posted screenshots showing the removed material. The Justice Department wiped Jan. 6 charge releases from its website, removing a public record built around about 1,600 defendants. Among the releases removed from the site were those concerning seditious conspiracy cases against members of the Proud Boys and Oath Keepers, far-right extremist groups, with the Justice Department, in an unopposed motion last month, asking a federal appeals court to vacate those seditious conspiracy convictions, a request that was granted Thursday.
The move represents an escalation in the Trump administration's revisionist approach to the events of January 6. Trump, on his first day back in office in January 2025, pardoned, commuted the prison sentences or vowed to dismiss the cases of all of the 1,500-plus people charged with crimes during the Capitol assault, including those convicted of attacking officers with makeshift weapons. The president not only commuted the sentences of many rioters, including those charged for violence, he also abruptly fired dozens of prosecutors who handled the cases. The administration has also announced a $1.8 billion "anti-weaponization fund" intended to compensate those claiming wrongful prosecution, with Acting Attorney General Todd Blanche not ruling out that rioters convicted of violence will be eligible for payouts, prompting bipartisan anger in Congress.
The removal of official legal documentation by a government department raises concerns about institutional integrity and the willingness of those in power to suppress inconvenient records. Citizens for Responsibility and Ethics in Washington said the deletion likely violated federal records law, citing 44 U.S.C. § 3106, which requires notice to the archivist when federal records are removed or deleted. On March 10, 2025, the National Archives opened an unauthorized-disposition case after the complaint. While the underlying court records remain public, and U.S. District Judge Paul Friedman, in a February 1, 2025 ruling, rejected Trump's claim that the prosecutions were a "national injustice" and ordered that a copy of the database be preserved on the federal court system's website, the scrubbing of DOJ communications signals a broader pattern of state capacity being used to reshape narratives around democratic accountability.
Generated at 2026-05-27 05:42 UTC