\n<\/p>\n
Dario Amodei, Chief Executive Officer of Anthropic, at the AI \u200b\u200bImpact Summit in New Delhi, India on Thursday, February 19, 2026. <\/p>\n
Ruhani Kaur Bloomberg | getty images<\/p>\n<\/div>\n<\/div>\n<\/div>\n
Anthropic announced a new artificial intelligence model on Thursday, cloud opus 4.7<\/a>Which the company said is an improvement over previous models, but “less comprehensively capable” than its most recent offering, the Cloud Mythos Preview.<\/p>\n Cloud Opus 4.7 is better at software engineering, follows instructions, gets real-world work done and is its most powerful generally available model, Anthropic said. But the model’s cyber capabilities aren’t as advanced as the Cloud Mythos Preview, which Anthropic introduced earlier this month to a select group of companies as part of a new cybersecurity initiative called Project Glasswing.<\/p>\n “We are releasing Opus 4.7 with safeguards that automatically detect and block requests that indicate prohibited or high-risk cybersecurity use,” Anthropic said in a release. \u201cWhat we learn from the real-world deployment of these security measures will help us work toward our ultimate goal of the widespread release of Mythos-class models.\u201d<\/p>\n Since its founding in 2021, Anthropic has spent years building its reputation as a firm that is more dedicated to safe and responsible AI deployment than rivals like OpenAI. <\/p>\n The launch of Project Glasswing has sparked a number of high-profile meetings between members of the Trump administration, tech CEOs, and bank CEOs about the security risks of powerful AI models. <\/p>\n<\/div>\n Anthropic doesn’t plan to make the Cloud Mythos preview generally available, but the company has said its goal is to learn how it can eventually deploy Mythos-class models at scale.<\/p>\n Thursday’s launch of Cloud Opus 4.7 follows Anthropic’s launch of Cloud Opus 4.6 in February. Anthropic said the new model outperforms Cloud Opus 4.6 across multiple use cases, including agentic coding, multidisciplinary reasoning, scaled tool use, and industry benchmarks for agentic computer use. a release<\/a>.<\/p>\n Anthropic said it used efforts to “partially mitigate” the cyber capabilities of Cloud Opus 4.7 during training. <\/p>\n The company encouraged security professionals who are interested in using the model for “legitimate cybersecurity purposes” to apply through the formal verification program.<\/p>\n Cloud Opus 4.7 is available across all of Anthropic’s cloud products, its application programming interface, and through cloud providers. Microsoft