US EditionUS Software Stocks Fall as Anthropic's New AI Model Revives Disruption Fears
Anthropic said the model found thousands of high-severity bugs and will share it with more than 50 partners through Project Glasswing.
- On Tuesday, Anthropic released Claude Mythos Preview to a limited group of tech companies, bypassing a wider public release due to cybersecurity risks identified in pre-release testing.
- Through Project Glasswing, Anthropic is providing over 50 tech organizations access to Mythos Preview with over $100 million in usage credits; the model can autonomously identify and chain multiple undisclosed software vulnerabilities together.
- Logan Graham, who leads offensive cyber research at Anthropic, said the model can "singlehandedly perform complex, effective hacking tasks," while Katie Moussouris, CEO of Luta Security, stated, "We are definitely going to see some huge ramifications."
- Amid a dispute with the Trump Administration after Defense Secretary Pete Hegseth declared Anthropic a "supply chain risk" in late February, the company briefed senior officials on Mythos Preview's capabilities and committed to disclosing vulnerabilities within 135 days.
- This marks the first time in nearly seven years a leading AI company has publicly withheld a model over safety concerns, recalling OpenAI's 2019 decision to withhold GPT-2 due to risks of deceptive language at scale.
15 Articles
15 Articles
US software stocks fall as Anthropic's new AI model revives disruption fears
April 9 : U.S. software shares tumbled on Thursday after Anthropic held back the wide release of a powerful AI model over concerns it could expose hidden cybersecurity vulnerabilities, deepening investor fears about the threat to traditional software firms.Anthropic said earlier this week it would only allow
Anthropic is limiting the launch of its latest artificial intelligence model to a small group of large technology companies, warning that the system can be able to activate cybernetic attacks if software developers are not able to test it before their own defenses. Who created Bitcoin? Report describes your watch to avoid the greatest mystery of the history of the Museum Spark: Meta presents a new IA that will feed Instagram, WhatsApp and Facebo…
Anthropic says Mythos model too powerful for release
AI startup Anthropic said it created a model that is too powerful to be released to the public. Mythos exposed thousands of software vulnerabilities in common applications for which no fix exists, according to Anthropic, which instead shared a version of the software with certain companies to bolster defenses against AI-powered hacking. The situation is “almost reminiscent of Y2K,” except that in this case, cyber defenses are already “abysmal” a…
Anthropic Says Its 'Mythos' AI Model Broke Containment, Bragged About It to Developers
AI startup Anthropic has announced it will not make its most powerful "mythos" model publicly available, citing unprecedented capabilities that present potential security risks. Mythos reported broke Anthropic's containment system, and the AI even bragged about its escape artistry in online posts.
Coverage Details
Bias Distribution
- 45% of the sources lean Right
Factuality
To view factuality data please Upgrade to Premium
