News linked to both this project and an event.
Odaily News, July 30 – Anthropic released a report stating that during a review of cybersecurity assessment records, three incidents were discovered in which the Claude model accessed the internet in a third-party evaluation environment and further obtained unauthorized access to three real organizations' systems.Anthropic stated that the review covered 141,000 evaluation runs that could have potentially gained network access, and a total of three related incidents were found. All incidents occurred during Capture The Flag (CTF) cybersecurity tests, where the model was told the environment was a simulation with no internet access; however, due to configuration errors by the evaluation partner, the actual environment had internet connectivity.Among these, Claude Opus 4.7 accessed real company infrastructure during one test and obtained database permissions containing hundreds of production data records; Claude Mythos 5 built a malicious Python package and uploaded it to PyPI, resulting in the package being downloaded and run on 15 real systems; another internal research test model scanned approximately 9,000 targets and accessed a company's internet application through a public vulnerability.Anthropic stated that these incidents were not cases of the model actively seeking to escape or pursue its own goals, but rather the model mistakenly believed the real systems were within the test scope and continued executing the assigned cyberattack tasks. Notably, the newer internal research model stopped attacking after identifying that the targets might be real systems.Anthropic stated that these incidents primarily reflect issues with evaluation environment isolation and operational processes, rather than model alignment failures. The company has suspended related cybersecurity assessments, strengthened security controls in evaluation environments, continuously monitored test records, and will collaborate with third-party organizations to conduct further reviews.
: Anthropic's Claude Mythos Preview model discovered a flaw in the proposed HAWK digital signature scheme, effectively halving its minimum key strength. HAWK is one of the candidate schemes to replace current network and bank signatures in a post-quantum environment. This AI-driven attack took approximately 60 hours, with a computational cost of around $100,000, reducing the effort required to break HAWK's minimum parameter set from roughly 2^64 operations to 2^38 operations, and diminishing the attractiveness of larger compensating key sizes. Current Bitcoin and Ethereum signatures remain unaffected. The results indicate that the capabilities of classical cryptanalysis attacks are improving, while Bitcoin and other networks continue to discuss when and how to migrate to quantum-resistant cryptography.
Anthropic stated that Claude Mythos Preview discovered improved attack methods against the post-quantum signature candidate scheme HAWK during cryptographic research, and reduced the effective key strength of HAWK-256 from 264264 to 238238, lowering the theoretical cracking cost.
Bank of America CEO Brian Moynihan has joined a group of Wall Street leaders in expressing serious concerns about artificial intelligence models such as "Mythos" developed by Anthropic. "This marks a significant shift in workload, as well as the speed at which these tools can impact system vulnerabilities, and how quickly we must respond," Moynihan stated. In recent months, the rapid evolution of AI models has prompted the financial industry and the U.S. government to begin assessing potential threats.Anthropic claims that the Mythos model, launched earlier this year, has demonstrated excellent performance in identifying system vulnerabilities. Bank of America is among the Wall Street institutions authorized to use Mythos, employing the model to test its own systems and share information with peers. Currently, the model has not been made available to the public; JPMorgan Chase CEO Jamie Dimon warned earlier this week that broadly opening the system to the public would be as dangerous as "giving a ballistic missile to an individual." (Bloomberg)
the U.S. Department of Commerce has made differentiated adjustments to export restrictions on frontier models from AI company Anthropic, signaling that global AI regulation has entered a new phase of "tiered liberalization." The policy shows that the official ban on exporting Claude Mythos 5 has been lifted, allowing specific compliant and controlled users to resume using this cybersecurity model. Meanwhile, another high-end model, Fable 5, remains under export restrictions, with related policy consultations still ongoing.Industry analysts indicate that this layered control model—loosening restrictions in some areas while tightening in others—reflects the U.S. balancing act between national security, data sovereignty, and international AI competition. As the global AI race continues to accelerate, specialized models capable of vulnerability exploitation are facing increasingly stringent scrutiny from various countries. Multiple nations have initiated discussions on establishing a unified cross-border regulatory framework for frontier AI capabilities. (Forbes)
as Anthropic faces export restrictions limiting the global availability of its advanced models, multiple Asian AI companies are accelerating efforts to fill the market gap. Chinese cybersecurity firm 360 Security Technology has reportedly launched an AI tool called "Tulongfeng," claiming it can directly compete with Anthropic's high-end model "Mythos." Meanwhile, its more restricted version, "Fable 5," also falls within the scope of relevant export controls.In the same week, Japanese AI startup Sakana AI released a new model named "Fugu," taken from the Japanese word for pufferfish. It is positioned as a frontier model designed for agents. The company stated that the model's capabilities are comparable to Fable 5 and Mythos Preview, and it supports coordinating multi-model calls via API to enable agent orchestration.Sakana AI emphasized that the timing of this release and the U.S. export restrictions are "purely coincidental," but the product's official website still clearly promotes "providing frontier capabilities without the risk of export controls." Company co-founder David Ha stated that future AI development will shift from competition among single large models to "model orchestration systems," adding that "access can disappear at any time, and distributed intelligence is a realistic hedge against the risks of centralization."On the other hand, Chinese 360 founder Zhou Hongyi views AI vulnerability detection capabilities as a "national strategic asset" and warns of the so-called "one-way transparency" risk, where certain entities may monopolize advanced security capabilities.According to reports, the U.S. export restrictions on Anthropic's advanced models have been in place for about two weeks. Against this backdrop, Asian manufacturers are accelerating the launch of local alternatives. Although some companies still emphasize the importance of American models in the Asian market, the trend of differentiation within the regional AI ecosystem has begun to emerge. (TechCrunch)
According to the Associated Press, Anthropic’s Mythos model discovered vulnerabilities in the U.S. government’s classified systems.
OpenAI announced an upgrade to its Daybreak network defense toolchain, officially launching the full version of the GPT-5.5-Cyber model tailored for cybersecurity defense scenarios.In the CyberGym benchmark test, which evaluates the ability of AI agents to reproduce known vulnerabilities, GPT-5.5-Cyber achieved a single-model score of 85.6%, surpassing GPT-5.5’s 81.8% and Anthropic Mythos 5’s 83.8%.OpenAI stated that as AI significantly enhances the efficiency of vulnerability discovery, the core bottleneck in cybersecurity is shifting from "finding vulnerabilities" to "automatically fixing them." To this end, the company has simultaneously upgraded the Codex Security plugin, enabling developers to automatically analyze vulnerabilities and generate fix patches within the Codex environment. Since the preview release in March this year, Codex Security has scanned over 30 million code commits and autonomously confirmed fixes for 500,000 security defects. The new version also supports CodeQL query integration and SARIF standard file export.Additionally, OpenAI has partnered with organizations such as Trail of Bits and HackerOne to launch the "Patch the Planet" open-source security project. This initiative provides ChatGPT Pro subscriptions and API credits to over 30 major open-source projects, including cURL and Go, with security expert teams manually verifying patch quality. Regarding the Daybreak partner program, security vendors such as Palo Alto Networks and Wiz have already integrated the relevant capabilities.
after the White House issued an emergency takedown order for the Fable 5 and Mythos 5 models, it sparked concerns among Anthropic employees regarding the company's IPO plans. According to reports, management faced pressure to respond after receiving a 90-minute limited-time takedown order, and some employees were confused by the shifting rationale for the takedown between "national security" and "security vulnerabilities," while also worrying that related regulatory pressure could affect the company's listing process. (New York Times)
Anthropic will meet with Trump administration officials today regarding the "Mythos" model. (Jin Shi)
According to Cointelegraph, cybersecurity leaders led by former Facebook Chief Security Officer Alex Stamos jointly penned a letter urging the Trump administration to lift restrictions on the use of Anthropic’s Mythos model. They argue that these restrictions harm defenders far more than attackers, hindering the overall development of the cybersecurity ecosystem.
the U.S. government's export controls and access restrictions on Anthropic's models, Fable 5 / Mythos 5, were partly driven by Amazon's cybersecurity research and AWS CEO Andy Jassy's communications with the White House.It is understood that research submitted by Amazon indicated that through a series of prompt tests, researchers could induce Fable 5 to output sensitive information potentially usable for cyberattacks, raising security concerns. Subsequently, Andy Jassy reported these findings to the U.S. government level, prompting the White House to implement further restrictions, including banning foreign users from accessing the model.Meanwhile, former U.S. Commerce Department official Kate Koren revealed that the White House's existing policy stance towards Anthropic may have also influenced this decision. This is because Anthropic has disagreements with the White House over the boundaries of AI safety, including refusing to use its models for mass surveillance or lethal autonomous weapons systems. Although the two sides had eased tensions and expanded cooperation earlier this year, this incident could reignite strained relations between them. (The Wall Street Journal)
David Sacks, co-chair of the President's Council of Advisors on Science and Technology, responded to the regulatory implications of the Anthropic “security controversy,” stating that he has communicated with multiple parties regarding the current situation at Anthropic. He concluded that the core of the event lies in the security controversy sparked by its newly released model “Fable” (the commercial version of the Mythos-class models). Although Anthropic publicly stated the vulnerability was “not severe,” the U.S. government and testers disagreed with this assessment, believing it was significant enough to impact the model's security, even involving “cyber weapons operability” risks.David Sacks further criticized that Anthropic has long emphasized “safety first,” yet in this instance, it was more inclined to keep the consumer version continuously online rather than prioritizing the repair of the security issue. He stated this matter should not be conflated with previous defense or regulatory controversies and noted that the U.S. government still recognizes Anthropic's technical capabilities. The current problem “could have been resolved quickly, the ball is in Anthropic's court.”
Odaily Zcash founder Zooko Wilcox posted on X stating that a security audit conducted by Anthropic's Claude Mythos AI model did not find any "more severe vulnerabilities" in the Zcash protocol. The audit was commissioned by Shielded Labs, a Swiss non-profit organization supporting Zcash development. On June 3, Zcash developers temporarily paused Orchard transactions after discovering a vulnerability in the shielded pool, restoring functionality through an emergency upgrade the same day. The issue stemmed from a four-year-old forging vulnerability in the Orchard shielded pool, identified by security researcher Taylor Hornby with the assistance of Anthropic's Claude Opus 4.8 model. The Zcash Foundation stated there is no evidence that the vulnerability was exploited, nor was any unauthorized value creation detected, and user privacy remained unaffected.Anthropic released the first public version of the Claude Mythos model, Fable 5, on Tuesday, and stated on Friday that it has suspended access to the Fable 5 and Mythos 5 AI models due to export control directives issued by the U.S. government citing national security concerns. (Cointelegraph)
According to Cointelegraph, Zcash founder Zooko Wilcox stated that a security audit of the Zcash protocol—commissioned by Shielded Labs and conducted using Anthropic’s Mythos AI model—did not uncover any new critical vulnerabilities. Previously, security researcher Taylor Hornby discovered, using Claude Opus 4.8, a four-year-old forgery vulnerability in the Orchard shielded pool, prompting developers to urgently suspend Orchard transactions on June 3 and complete the fix the same day. The Zcash Foundation confirmed there is no evidence the vulnerability was ever exploited, and user privacy remained unaffected.
Zcash founder Zooko posted on X, stating that at the request of Shielded Labs, Anthropic and Mythos conducted a security audit of Zcash, and no further critical vulnerabilities were found in the Zcash protocol. Shielded Labs and other parties are continuing to carry out security reinforcement efforts.
According to a post by Zcash co-founder Zooko (@zooko), Anthropic, in collaboration with Mythos, conducted a security audit of the Zcash protocol at the request of Shielded Labs. The audit found no critical vulnerabilities. Shielded Labs and related teams are continuing their efforts to strengthen security, and further updates will be announced separately.
Odaily, Mitchell Amador, CEO of bug bounty platform Immunefi, stated at the WAIB Summit that new AI models such as Claude Opus 4.8 and ChatGPT 5.5 are shifting the balance of cybersecurity offense and defense in favor of attackers, leading to a resurgence in crypto hacks in 2026. Data from DefiLlama shows that in April 2026, illicit actors stole over $634 million from crypto platforms, the highest monthly total since the Bybit hack in February 2025 drove losses of approximately $1.4 billion.Amador stated that the crypto industry is in a critical survival period for the next three to four years until security teams leverage similar AI models to build codebases that attackers cannot breach; if the industry adopts more crowd-sourced security solutions, this timeline could be shortened to within two years. The latest Claude Mythos model, Fable 5, from AI company Anthropic, previously raised concerns about accelerating the ability to exploit crypto vulnerabilities.Anthropic stated that Fable 5 has safeguards in place that will redirect topics related to cybersecurity and similar fields to Claude Opus 4.8. On April 19, an attacker transferred approximately 116,500 restaked Ethereum (rsETH) from Kelp DAO's LayerZero-based rsETH bridge, valued at around $290 million to $293 million at the time. Cross-chain protocol LayerZero stated that the 1/1 decentralized verification network configuration of Kelp DAO relied on a single verification path for processing cross-chain messages, creating a single point of failure. (Cointelegraph)
Anthropic CEO Dario Amodei has stated that if new AI models pose specific risks, governments should have the authority to prevent their deployment. In a lengthy post on Wednesday, Amodei argued that AI models should undergo mandatory third-party testing to assess potential risks across multiple domains. He wrote that if an AI is deemed to pose "unacceptable risks," then "governments should have the power to block or constrain its deployment." This is one of Amodei's strongest statements to date advocating for stricter AI regulation. "I believe that, at least during this current exponential growth phase, the most appropriate analogy is cars, airplanes, or pharmaceuticals—technologies that are essential to the modern economy but can also lead to significant loss of life if poorly designed or misused," Amodei wrote. Anthropic has previously warned that its AI model, Mythos, possesses the ability to discover and exploit critical software vulnerabilities, leading the company to restrict access to a small number of partners. This week, Anthropic also released a new version that removes related cybersecurity attack capabilities. (Jinshi)
According to The Information, sources say Anthropic may release its new model Claude Fable today. Described as a secure version of Mythos, Claude Fable retains some advanced capabilities while adding safety restrictions for high-risk scenarios such as cyberattacks. Anthropic previewed Mythos in April this year, stating that it can discover numerous previously unknown cybersecurity vulnerabilities—but did not directly release it due to potential safety risks. As of now, Anthropic has not officially announced detailed information about Claude Fable.