anthropic

According to the Associated Press, Anthropic's Mythos model has discovered vulnerabilities in the U.S. government's classified systems.

Micron Technology and Anthropic announced a new agreement, covering collaboration in memory and storage AI architecture design, supply and demand coordination, Micron's internal adoption of Claude, and Micron's strategic investment in Anthropic's Series H financing.

Coinbase Markets announced the launch of pre-IPO perpetual contracts for Anthropic and OpenAI, with contract codes ANTHROPIC-PERP and OPENAI-PERP, respectively.The official statement said that the pre-IPO perpetual contract market for both contracts will open for trading on Coinbase on June 22, 2026, at 11:00 UTC (19:00 Beijing time) or later.

According to Andrew Curran, the next-generation Mythos model from Anthropic, which has stronger capabilities (named either Mythos 5.1 or Mythos 6), has completed training. It is currently uncertain whether this model will be publicly released or kept for internal use only to accelerate the development of subsequent technologies.Analysis indicates that although some cutting-edge models like Fable 5 or Mythos 5 face restrictions or scrutiny regarding public release, this has not slowed down the research and development pace of top AI laboratories. In the face of fierce competition from open-source models represented by GLM-5.2, leading AI companies still need to continuously invest in and train more powerful systems to maintain their commercial and technological leadership.

Nansen CEO Alex Svanevik posted on platform X that, in hindsight, if Binance founder CZ had completed the acquisition of FTX back then, the potential asset structure could have changed significantly, possibly indirectly holding about 8% exposure in Anthropic, about 5% exposure in the AI programming tool Cursor, and some investment rights related to SpaceX.It is reported that CZ disclosed the intention to acquire in November 2022, but announced the abandonment due to issues discovered during due diligence that were beyond control, after which FTX filed for bankruptcy protection.

According to a report by Business Insider, Anthropic CEO Dario Amodei recently discussed the competitive relationship with OpenAI and its CEO Sam Altman in an exclusive interview. Amodei stated that he feels "calm" about the current competitive landscape between the two companies and candidly pointed out that their visions differ and there is a lack of trust. He said, "When you don't have a common vision and don't trust each other, arguing is meaningless," and emphasized that both sides should pursue their own paths, ultimately leaving the market and public opinion to judge the outcome.In addition, Amodei proposed that the AI industry needs "trustworthy participants" to jointly establish standards, thereby forcing other untrustworthy peers to comply with regulations. He specifically mentioned that he has known Google DeepMind CEO Demis Hassabis for 15 years, and the two have maintained close cooperation and trust in computing power procurement and safety concepts, thereby refuting the widespread rumor of "mutual distrust" in the AI field.

According to the Financial Times, informed sources revealed that JPMorgan Chase has stopped its Hong Kong employees from accessing Anthropic's AI models. Currently, the bank's Hong Kong employees can no longer call models like Claude from the internally approved large language model dropdown list.It is reported that this decision is primarily based on the specific wording of the usage terms in the licensing agreement between JPMorgan Chase and Anthropic. According to the relevant terms, the usage scope of Anthropic's models excludes the Greater China region (including Hong Kong). Earlier this year, Goldman Sachs also implemented similar access restrictions for its Hong Kong employees based on a strict interpretation of these terms.Although international institutions previously could usually circumvent regional restrictions by signing global contracts and hosting data overseas, compliance actions among Wall Street banks are tightening as the U.S. government and regulatory agencies increasingly scrutinize the overseas use of AI technology. Anthropic has previously made it clear that its Claude model has never been officially "supported" in Hong Kong.

According to a report by The Information, Broadcom announced its support for the Anthropic computing power project participated by Apollo and Blackstone.On the surface, this deal appears to be a financing arrangement for AI computing infrastructure, but industry insiders believe that its core lies in indirectly driving the long-term demand growth for Broadcom's custom AI chips by promoting large-scale AI computing deployments.

According to the Wall Street Journal, the U.S. government's export controls and access restrictions on Anthropic's models Fable 5 / Mythos 5 partly stem from Amazon's cybersecurity research and AWS CEO Andy Jassy's communications with the White House.It is reported that the research submitted by Amazon indicated that through a series of prompt tests, researchers were able to induce Fable 5 to output sensitive information that could potentially be used for cyberattacks, raising security concerns. Subsequently, Andy Jassy reported the relevant findings to the U.S. government, prompting the White House to take further restrictive measures, including banning foreign users from accessing the model.At the same time, former U.S. Department of Commerce official Kate Koren revealed that the White House's existing policy stance on Anthropic may also have influenced this decision-making judgment, as there are differences between Anthropic and the White House on AI safety boundary issues, including the refusal to use the model for large-scale surveillance or lethal autonomous weapon systems. Although both sides had eased tensions and expanded cooperation earlier this year, this incident may once again exacerbate the strained relationship.

David Sacks, the head of AI and crypto affairs at the White House, stated that although the commercial version of the Mythos series model Fable released by Anthropic this week includes safety barriers, once those barriers are bypassed, users will be able to access Mythos's advanced cyber attack capabilities. Sacks pointed out that Anthropic had previously described Mythos as a "cyber weapon" that requires regulation, so fixing the related vulnerabilities should have been its responsibility.Sacks mentioned that a partner trusted by both Anthropic and the U.S. government discovered a jailbreak method to bypass the safety barriers while testing Fable. The U.S. government subsequently requested Anthropic CEO Dario Amodei to fix the vulnerabilities or take the model offline, but this request was refused. Anthropic stated in a declaration that the vulnerability is "not serious," a claim that contradicts the judgment of the U.S. government and relevant partners.Sacks stated that Anthropic has always emphasized that safety should be prioritized, but this time it prioritized maintaining consumer-grade model services. In response, the U.S. government reluctantly imposed export control measures on Anthropic and hopes that Anthropic will resolve the safety issues as soon as possible to lift the related restrictions and restore the full release of Fable.Sacks also denied that this action is related to previous disputes between the U.S. Department of Defense and Anthropic, stating that the government recognizes Anthropic's technological capabilities and believes that the current issues can be resolved relatively easily, with the initiative currently in Anthropic's hands.

According to an official statement from Anthropic, on June 11 local time, the U.S. government issued an export control directive citing national security, requiring the suspension of access to Fable 5 and Mythos 5 for all foreign nationals. Anthropic was forced to shut down the aforementioned two models for all users.The government believes it has discovered a "jailbreak" method to bypass the protective mechanisms of Fable 5, but Anthropic, after review, pointed out that the related vulnerability is relatively minor and that similar capabilities exist in other publicly available models such as OpenAI GPT-5.5, which does not constitute an exclusive risk to Mythos.Anthropic stated that it will comply with the government directive but clearly opposes taking down commercial models that have been deployed to hundreds of millions of users based on the discovery of a single narrow-range jailbreak. They warned that if this standard is widely applied in the industry, it would effectively block all new deployments of cutting-edge models. The company is currently actively communicating with the government to restore access as soon as possible.

According to Reuters, Anthropic has signed more than 12 preliminary leasing agreements for data centers in the United States, with a total capacity exceeding 1 gigawatt. It is reported that Anthropic executives are discussing with Google about providing financial guarantees for these lease payments, while Apollo Global Management and Blackstone are providing private credit as part of a broader financing structure.Google's Gemini directly competes with Anthropic's Claude, but Google has committed up to $40 billion in investment to Anthropic and is participating in joint server chip design. Last month, Anthropic completed a $65 billion financing round with a valuation of $965 billion, surpassing OpenAI to become the AI company with the highest implied market value, and has secretly submitted a U.S. IPO application.Previously, Anthropic mainly relied on cloud service providers like Google Cloud for computing power, and building its own data centers aims to reduce dependence on competitors. In addition, Anthropic pays SpaceX (also a competitor through xAI) $1.25 billion per month for computing power. Neither Google nor Anthropic has publicly responded to specific details.

Alliance co-founder Imran Khan posted on X: "There is a project team within Alliance composed of former team members from Anthropic, exploring a very interesting entry point that focuses on the intersection of cryptocurrency, trading, and AI. The imaginative space and potential market in this direction are much larger than they appear on the surface. We will share more details soon, stay tuned."

OpenAI is considering significantly lowering product prices to respond to the potentially escalating fierce competition with Anthropic.Reports indicate that OpenAI expects competition in the AI services market to intensify further and may evolve into a price war. Meanwhile, some corporate clients have begun to show greater sensitivity to the current pricing levels of AI products.Insiders say that against the backdrop of corporate users' ongoing focus on AI usage costs, the industry pricing system may face restructuring, and competitors may also be forced to adjust their pricing strategies. Analysts believe that as the generative AI industry enters a new phase of commercial competition, price and control of inference costs are becoming important factors for major manufacturers vying for market share.

Developers can officially call the latest Claude Fable 5 model released by Anthropic through the B.AI platform. As Anthropic's strongest coding model to date, Claude Fable 5 is designed for challenging engineering scenarios such as legacy system migration, complex bug troubleshooting in production environments, and long-cycle asynchronous development. It excels in code generation, logical reasoning, and problem-solving. This model further enhances long-context understanding and task planning capabilities, able to handle complex development processes that last for hours or even days. Currently, the API is fully open, supporting flexible integration and invocation; the web-based dialogue function will also be launched on the B.AI platform soon, providing developers with a more convenient and smooth interactive experience.

According to Fortune, Goldman Sachs and Morgan Stanley are competing for the lead underwriter position for the future IPOs of OpenAI and Anthropic. Previously, Goldman Sachs secured the lead underwriter role for the SpaceX IPO, which is expected to take place on June 12. Both OpenAI and Anthropic's IPOs are expected to raise at least $60 billion, totaling over $120 billion.Jay Ritter, an IPO expert from the University of Florida, stated that the investment bank that secures the lead underwriter position will determine how much IPO stock allocation each fund receives, thus potentially earning far more than other underwriters. Ritter noted that institutional investors often pay a significant amount of "soft dollars" to lead underwriters to secure more IPO allocations, which refers to the portion of trading commissions that exceeds the actual execution costs.If it is still uncertain whether Goldman Sachs or Morgan Stanley will secure the lead underwriter position for OpenAI and Anthropic, funds may simultaneously direct more trading commissions to both investment banks to increase their chances of receiving allocations. For example, in the case of the SpaceX IPO, if the fundraising scale reaches $86 billion, the underwriting fee would be approximately $600 million. However, if the stock price rises by 20% on the first day, the paper gains for IPO investors would exceed $17 billion, and Ritter estimates that at least 30% of that could flow back to the investment bank in the form of soft dollars, about $5 billion, far exceeding the underwriting fee itself.If both OpenAI and Anthropic see a 20% increase on their first day of trading, the paper gains for investors would reach $24 billion. Based on the same ratio, investment banks could potentially receive over $7 billion in soft dollars, most of which would flow to the lead underwriter. Currently, both OpenAI and Anthropic have secretly submitted draft registration statements, but the listing dates have not yet been determined.