115 primary articles · 106 secondary mentions
Problem This paper addresses the gap in understanding the adversarial robustness of large language models (LLMs), specifically Anthropic’s Fable 5 and Opus 4.8, against automated jailbreak attacks. The study is...
In a significant shift, Anthropic has decided to reverse its planned billing overhaul for the Claude Agent SDK just ahead of its launch. This decision comes at a critical time...
The U.S. government’s recent decision to ban certain AI models developed by Anthropic has sparked significant discussion about the implications of regulatory interference in the tech sector. This move, which...
The U.S. government is in ongoing discussions with Anthropic, a leading AI company, regarding the development of unhackable large language models (LLMs). This request, driven by national security interests, poses...
Anthropic’s decision to shut down its AI models, Fable 5 and Mythos 5, has triggered a significant discussion about Europe’s technological sovereignty and its ability to develop independent AI infrastructure....
Anthropic, a prominent player in the AI sector, has recently underscored its unwavering commitment to safety in artificial intelligence. This declaration comes at a crucial time as the industry grapples...
In a significant development for the AI landscape, the White House recently ordered the shutdown of Anthropic’s Fable model, a move reportedly spurred by Amazon and five other companies. This...
In a significant move, Anthropic has suspended access to its latest AI models, raising alarms among tech leaders in India about the implications for the country’s burgeoning AI sector. This...
Anthropic’s latest AI model, Claude Fable 5, has achieved a significant milestone by outperforming OpenAI’s GPT-5.5 on FrontierMath’s toughest problems, boasting an impressive accuracy rate of 88%. This development is...
Anthropic, a prominent AI company, is currently grappling with a government decision to recall its most powerful AI model, a move that has sparked frustration within the organization. This situation...
A new survey conducted by Anthropic highlights growing concerns among Americans regarding the impact of artificial intelligence on employment and cognitive autonomy. With 52,000 participants, the findings reveal that 64%...
Anthropic has unveiled its latest AI model, Claude Fable 5, which has garnered attention for achieving a score of 64.9 on the Artificial Analysis Intelligence Index. This performance marks a...
Anthropic, an AI company known for its development of advanced models, is currently facing pushback from customers, partners, and investors regarding its Mythos model. This situation is significant as it...
Problem This work addresses the limitations of existing recursive language models (RLMs) and coding agents, particularly in their ability to handle long-context reasoning effectively. While RLMs have demonstrated the utility...
Dario Amodei, co-founder of Anthropic, has released a thought-provoking essay that positions artificial intelligence as a strategic weapon in the hands of nation-states. This perspective is particularly relevant as global...
Problem This work addresses the gap in understanding the internal mechanisms of reward models used in reinforcement learning from human feedback (RLHF), particularly the conflicting objectives of helpfulness and harmlessness....
Anthropic has recently admitted to making a “wrong tradeoff” in its approach to AI research, which involved throttling the capabilities of its rivals. This acknowledgment comes at a critical time...
Dario Amodei, the CEO of Anthropic, is making waves in the AI industry with a notably lean leadership structure—he has just one direct report. This unconventional approach is drawing attention...
Anthropic’s latest AI model, Fable, has drawn criticism from cybersecurity researchers who argue that its stringent guardrails hinder effective security work. This backlash highlights a growing tension between the need...
Anthropic has unveiled Claude Fable 5, the inaugural model in its new Mythos class, which is designed to enhance AI capabilities while maintaining stringent safety measures. This launch is particularly...
A significant development in the AI landscape has emerged with the introduction of Fable 5, the public version of Mythos. This release is particularly noteworthy as it showcases advanced capabilities...
In a significant move within the venture capital landscape, Justin Ernest, founder of Sabertooth VC, has successfully raised nearly $500 million over the past year to invest in high-potential startups....
Anthropic has launched Fable 5, a software tool designed to enable users to create video games effortlessly with just a click. This development is particularly significant as it caters to...
Anthropic has introduced two new AI models, Claude Fable 5 and Mythos 5, which promise substantial improvements in coding and scientific research capabilities. This release is particularly significant as it...
Anthropic has unveiled Claude Fable 5, marking the first time its advanced Mythos-class model is accessible to the public. This release is significant as it comes amid growing concerns over...
Problem This preprint addresses the gap in understanding how large language models (LLMs) adapt math word problems across diverse languages and cultural contexts. While LLMs are increasingly employed for personalized...
Problem This work addresses the gap in understanding the effects of reinforcement learning from human feedback (RLHF) on the alignment of large language models, specifically regarding the retention of partisan...
The U.S. Department of Defense is currently in a race to replace Anthropic’s AI model, Claude, which has been deemed “too safe” for military use. This decision comes amid growing...
Notion, the popular productivity platform, has successfully restored access to Anthropic’s AI services after a brief disruption that affected users. This incident underscores the growing reliance on AI technologies in...
In a significant move within the AI industry, Anthropic has successfully recruited Clive Chan, OpenAI’s second-ever chip engineer, who played a crucial role in the development of the company’s custom...
The National Security Agency (NSA) is reportedly leveraging Anthropic’s Mythos AI model for offensive cyber operations aimed at China and Iran. This development underscores the increasing intersection of artificial intelligence...
In a pivotal moment for Anthropic, co-founder Daniela Amodei has publicly addressed concerns regarding the financial viability of artificial intelligence as the company gears up for its initial public offering...
Lovable, a rising player in the AI landscape, has announced a significant multiyear partnership with Google Cloud. This collaboration is poised to expand Lovable’s operational footprint on the cloud platform...
Problem This work addresses the gap in AI agent recovery mechanisms when encountering API validation errors. Traditional APIs provide verbose error messages that lack actionable guidance for agents, necessitating external...
Anthropic has significantly scaled its Project Glasswing initiative, now partnering with 150 organizations across 15 countries to tackle critical software vulnerabilities. This expansion is particularly timely as the demand for...
Anthropic is significantly scaling its Project Glasswing initiative, which focuses on identifying and mitigating security vulnerabilities, by extending access to its AI model, Mythos, to 150 organizations across 15 countries....
Anthropic has taken a significant step by confidentially filing a draft registration for an initial public offering (IPO) with the US Securities and Exchange Commission (SEC). This move comes as...
Anthropic, the AI safety and research firm, has taken a notable step by filing confidentially for an initial public offering (IPO). This move comes at a time when the AI...
Anthropic has made a significant move by banning the use of AI tools during its job interviews, opting instead for a rigorous assessment process that includes up to five rounds...
A recent study by Anthropic reveals a significant gender disparity in the use of AI coding agents among social science researchers, with male researchers utilizing these tools more than twice...
In a striking example of unchecked AI expenditure, an unnamed company reportedly spent $500 million on Claude licenses in just one month due to a lack of usage limits. This...
Anthropic, the AI company behind the Claude language model, has successfully raised $65 billion in its Series H funding round, bringing its valuation to an impressive $965 billion. This significant...
Anthropic has unveiled Claude Opus 4.8, a new AI model that surpasses OpenAI’s GPT-5.5 and Google’s Gemini 3.1 Pro in various performance benchmarks. This release is significant as it highlights...
Anthropic has successfully secured $65 billion in its Series H funding round, bringing its post-money valuation to an impressive $965 billion. This significant capital infusion positions the AI startup on...
Problem This preprint addresses the gap in understanding the role of AI agents in scientific software development, particularly in the context of physicist-supervised AI. The study investigates the limitations of...
Problem This paper addresses the limitations of static reward models in Reinforcement Learning from Human Feedback (RLHF), which struggle to generalize across diverse and unseen human preference domains. Existing multi-reward...
Anthropic has unveiled Opus 4.8, featuring a groundbreaking tool called Dynamic Workflows designed to enhance the coordination of multiple subagents. This release is significant as it represents a step forward...
Elon Musk’s recent comments on the compute deal between xAI and Anthropic have sparked debate regarding the duration and stability of the partnership. While Musk suggests that the agreement is...
Problem This paper addresses the gap in stance detection within prediction markets, specifically focusing on trader commentary, which is characterized by extreme brevity, domain-specific language, and significant class imbalance (only...
Anthropic’s Claude Mythos has reportedly solved the long-standing Erdős unit-distance conjecture, a problem that OpenAI recently tackled. This development, announced by engineer Sholto Douglas, highlights the rapid advancements in AI’s...
