Anthropic launches top-tier AI safety model, currently limited to tech giants for trial use; cybersecurity stocks generally rise.

```

Anthropic is handing a more powerful undisclosed AI model over to tech giants for testing, in order to address cybersecurity threats that advanced AI systems may pose. This move has triggered an immediate reaction in the market, with cybersecurity stocks rising.

On Tuesday, April 7, Anthropic announced the launch of an industry joint project called "Project Glasswing." Amazon, Apple, Microsoft, Cisco and other companies will gain access to its undisclosed new model, Mythos, for examining vulnerabilities in their own products and sharing findings with industry peers.

Concerned about hackers using the model to launch cyberattacks, Anthropic stated that it currently has no plans to release Mythos to the public, and will use feedback from Project Glasswing to set safety guardrails for this technology.

Boosted by this news, global cybersecurity stocks rose broadly on Tuesday. The Global X Cybersecurity ETF gained 0.9%, marking its sixth consecutive day of gains.

Palo Alto Networks rose 4.9%, CrowdStrike was up 6.2%, Zscaler increased 1.8%, and Fortinet climbed 1.7%. Analysts believe Anthropic’s move clearly demonstrates the boost in demand for cybersecurity software driven by AI technology.

Project Glasswing: Giving Defenders a Head Start

The core logic of Project Glasswing is to identify and patch potential cybersecurity vulnerabilities before powerful AI models are deployed more widely. Participating companies will proactively use the Mythos model to find defects in their products and share their findings within the industry.

Newton Cheng, who leads Anthropic’s advanced red team cybersecurity efforts, stated:

We believe this is not just Anthropic’s problem—it’s an industry-wide issue. Both private enterprises and government bodies need to take it seriously. What we’re doing with Glasswing is giving defenders a first-mover advantage.

Analysts believe this reflects growing concerns in the tech industry. As AI model capabilities improve, criminals and hackers may use such tools to scan source code for vulnerabilities and breach cyber defenses.

Anthropic’s competitor OpenAI has also highlighted the growing cyber capabilities of its models, and launched pilot projects aiming to "put tools in defenders’ hands first."

Anthropic said it has communicated with U.S. government officials about Mythos’s safety-related capabilities, but declined to specify which agencies were involved.

Newton Cheng noted the company’s current work involves cooperation with the U.S. Cybersecurity and Infrastructure Security Agency (CISA) and the National Institute of Standards and Technology (NIST).

Analysts: Industry Threat Environment Reaches a Turning Point

Wall Street analysts generally interpret Anthropic’s move as a positive signal for the cybersecurity industry, and believe this event redefines the relationship between AI companies and security vendors.

Jefferies Financial Group believes this is a “clear signal of cooperation, not competition” for Palo Alto Networks and CrowdStrike, and indicates that “the threat environment is reaching a turning point”, giving both of these platform companies “strong excess return potential in the emerging AI era.”

Financial services firm Stephens commented that the news “validates the severity of more powerful AI models at the cybersecurity level and the necessity for cybersecurity vendors and key enterprise software providers to collaborate and jointly defend against complex threats.”

Stephens predicts enterprises will increasingly rely on trusted cybersecurity vendors to protect increasingly complex heterogeneous AI environments.

Private wealth management firm Bernstein pointed out from another angle that “Anthropic clearly highly values code security and the issue of its models being maliciously exploited”, and believes this “basically doesn’t pose resistance for cybersecurity vendors, and could even be a tailwind.”

Risk Disclosure and DisclaimerThe market carries risk and investment requires caution. This article does not constitute personal investment advice, nor does it take into account the special investment objectives, financial conditions, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article fit their particular situation. Any investment based on this is at your own risk. ```