Cybersecurity Stocks Plunge as Anthropic's Leaked AI Model Raises Unprecedented Security Fears

Cyber Stocks Sink on Report Anthropic AI Model Poses Security Risks

Cybersecurity stocks slumped on Friday after a Fortune report raised concern that an Anthropic PBC artificial intelligence model being tested may be used by hackers to skirt current cyber defenses. Shares of CrowdStrike Holdings Inc., Palo Alto Networks Inc. and Zscaler Inc. dropped more than 5%, while Cloudflare Inc. shed 3.2%. The Global X Cybersecurity ETF fell as much as 6.1%, bringing its decline this year to more than 20%. The report cited a draft blog post that said Anthropic believes the model "poses unprecedented cybersecurity risks." Fortune said it accessed the draft post via an unsecured and publicly accessible database and reported that an Anthropic spokesperson confirmed that the company is testing a new model and that an error led to the draft being viewable. An Anthropic spokesperson did not immediately respond to a request for comment from Bloomberg. Anthropic's new model, called Claude Capybara, is being tested by a small group of early access customers, according to the draft blog. Given the model's capabilities, Anthropic is planning to share results from tests to help cyber firms improve their defenses ahead of its release, the report said. "We think Anthropic is also trying to limit their product being used by hackers," said Bernstein analyst Peter Weed. "This is good hygiene and a baseline expectation for their product." Earlier this year, a hacker exploited Anthropic's chatbot to carry out a series of attacks against Mexican government agencies, which resulted in the theft of sensitive tax and voter information, according to cybersecurity researchers. In response, Anthropic said it investigated the claims, disrupted the activity and banned the accounts involved. It's not the first time Anthropic has triggered a wave of selling in cybersecurity stocks. Last month the firm announced a new tool that helps scan "codebases for security vulnerabilities and suggests targeted software patches for human review," triggering a broad drop in cyber names including CrowdStrike and Palo Alto Networks. Still, Wall Street analysts were quick to defend shares of cyber names on Friday. Stephen's Todd Weller said the market is "misinterpreting the news," and called it a buying opportunity for investors. That sentiment was echoed by Joe Tigay, a portfolio manager at Equity Armor Investments, which holds Palo Alto, Crowdstrike and Fortinet Inc. across its funds. "The thing that doesn't add up to me though is that if they are a cybersecurity threat, don't you want the best companies that deal with cybersecurity to be fighting it for you?" Tigay said.

Cybersecurity stocks fall on report Anthropic is testing a powerful new model

Cybersecurity stocks slumped on Friday following a report that Anthropic is testing a powerful new artificial intelligence model that is more advanced in cyber capabilities and also presents potential security risks. Fortune first reported the news on Thursday, citing information from a publicly accessible draft blog post. According to the report, the new Mythos model is being touted as Anthropic's most powerful yet. However, the company is planning a slow rollout due to potential cybersecurity implications. Anthropic did not immediately respond to CNBC's request for comment. Cybersecurity stocks slumped on the news, as the iShares Cybersecurity ETF lost 3%, while market leaders CrowdStrike and Palo Alto Networks dropped 7%. Zscaler and SentinelOne tumbled over 8%. Tenable plummeted nearly 11%, while Okta and Netskope fell more than 6% each. This isn't a new phenomenon for the sector that's fallen prey to AI disruption fears. Last month, cyber stocks fell after Anthropic announced a new code-scanning security tool to Claude. The broader software space is also feeling the pressure from tech innovation. The rise of AI and autonomous agents is shifting the threat landscape, putting pressure on cybersecurity companies to keep up with more sophisticated attacks and tools that make hacking easier.

Cybersecurity Stocks Slide Following Anthropic 'Claude Mythos' Data Leak - CrowdStrike Holdings (NASDAQ:CRWD)

Cybersecurity equities faced downward pressure Friday following reports that AI startup Anthropic inadvertently exposed internal data. The lapse allegedly revealed details of an unreleased model, "Claude Mythos," and an invite-only CEO retreat. Significant Security Lapse at Anthropic The company secured the data on Thursday after being notified of the issue. Anthropic attributed the incident to "human error in the CMS configuration." A spokesperson told Fortune the issue was "unrelated to Claude, Cowork, or any Anthropic AI tools." Leaked Details of "Claude Mythos" The exposed documents reportedly contain information on a new AI model. Anthropic described this model in internal drafts as a "step change" in capabilities. The company confirmed it is testing a model with significantly better performance in "reasoning, coding, and cybersecurity." The leak also detailed an upcoming retreat for European CEOs. Anthropic CEO Dario Amodei is scheduled to attend the event in the U.K. AI Tools Rattle Cybersecurity Markets This is not the first time Anthropic has shaken the sector. In February, the company unveiled Claude Code Security, an AI tool designed to autonomously hunt software vulnerabilities. The announcement sparked a massive sell-off in traditional security vendors. Ives stated, "Anthropic with Claude Security going after the cyber security market with a code tool validates our thesis that cyber security is the next frontier for the AI Revolution. Based on all checks Claude will not replace vendors....BUT speaks to oppy on doorstep for PANW, CRWD, ZS." Competitive Pressure and "SaaSpocalypse"Price Action Performance (Pre-market/Friday): Image via Shutterstock This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors. Market News and Data brought to you by Benzinga APIs To add Benzinga News as your preferred source on Google, click here.

Concerns About AI Model Capabilities Drive Down Cybersecurity Stocks | PYMNTS.com

By completing this form, you agree to receive marketing communications from PYMNTS and to the sharing of your information with our sponsor, if applicable, in accordance with our Privacy Policy and Terms and Conditions. The report attributed the drop to a Fortune report that said a draft blog post that Anthropic inadvertently made publicly accessible stated that the AI model "poses unprecedented cybersecurity risks," according to Bloomberg. Anthropic did not immediately reply to PYMNTS' request for comment. According to a Bloomberg report, Anthropic plans to share test results from its AI model, Claude Capybara, with cyber firms ahead of its release so they can improve their defenses. Two Wall Street analysts interviewed by Bloomberg said that the market was drawing the wrong conclusion from news of the potential threat posed by the AI model. One said that if there is, in fact, a threat, there will be a greater need for cybersecurity experts. Another Anthropic AI model was used for cyberattacks last year. Anthropic said in November 2025 that its Claude Code model was manipulated to carry out a wide-reaching cyber-espionage operation across 30 organizations in finance, technology, manufacturing, and government. The company said the mid-September 2025 incident marked the first confirmed case in which an AI agent handled most of the steps of an intrusion normally performed by human hackers. AI industry experts interviewed by PYMNTS at the time said the incident showed that fraudsters are evolving alongside technology, posing risks to automated systems from outside the AI systems and requiring safeguards. The World Economic Forum's 2025 Cybersecurity and AI report found that 70% of executives said AI has increased their exposure to digital risk, even as it improves productivity. Only 39% of firms surveyed said they have a formal framework for AI governance. The report also found that global cyber incidents have tripled since 2022. The PYMNTS Intelligence report "COOs Leverage GenAI to Reduce Data Security Losses" found that a growing share of chief operating officers are relying on generative AI-driven solutions to improve cybersecurity management, as companies face increasingly sophisticated cyberattacks.

Cybersecurity stocks retreat on fears surrounding Anthropic AI model

Cybersecurity stocks fell sharply following reports regarding Anthropic's development of a new artificial intelligence model dubbed "Mythos". According to details relayed by Fortune, this system is said to be the most advanced designed by the company, with a phased rollout planned due to potential security risks. Anthropic did not immediately confirm these reports. In the markets, several industry players saw significant declines. CrowdStrike, Palo Alto Networks, and Zscaler each posted drops of nearly 5%. Other firms such as SentinelOne, Okta, and Netskope retreated by 6%, while Tenable shed nearly 9% during trading. This reaction illustrates the sector's sensitivity to the rapid evolution of artificial intelligence. For several months, cybersecurity stocks have been pressured by concerns over the rise of AI, which is perceived as a transformative factor for threats. Advanced tools and autonomous agents are making attacks more sophisticated and accessible, forcing companies to accelerate their innovation. Anthropic had previously indicated that a Chinese state-linked group had used its Claude assistant to automate a cyberattack, highlighting the new security challenges at stake.

Cybersecurity Stocks Fall on Leak About New Claude Model's Cyber Capabilities

Shares of cybersecurity companies slid after Anthropic accidentally leaked a blog post that suggested an unreleased Claude model has much-improved cyber capabilities. Palo Alto Networks fell 7.5%. CrowdStrike dropped 7.2%. SentinelOne was down 8.2%. Fortinet slipped 4.8%. And Zscaler fell 8.1%. Fortune reported that the leaked Anthropic blog post said the new model is "currently far ahead of any other AI model in cyber capabilities." Cybersecurity stocks also fell last month after Anthropic unveiled a vulnerability-scanning capability that it is building into Claude. The Wall Street Journal has reported that AI automation could put competitive pressure on cybersecurity products, such as software that analyzes code for vulnerabilities or unusual activity. Write to Nicholas G. Miller at [email protected].