Anthropic CEO Dario Amodei has called on artificial intelligence companies to slow the pace at which they develop increasingly capable AI models, arguing that the industry needs more time to strengthen safeguards against potential misuse.Amodei made the appeal in an essay published on Friday, a day after Anthropic released a threat intelligence report detailing cases in which people and organisations allegedly used its Claude AI models for activities including cyberattacks, surveillance, weapons development and fraud, news agency Reuters reported.Amodei said he was not calling for AI development or model training to stop. Instead, he argued that companies should make sure safety and alignment measures keep pace with rapidly improving capabilities, with independent experts given a role in evaluating those safeguards.What did Anthropic’s CEO propose?Amodei proposed a three-step approach to slowing the effective pace of AI development.One of his proposals is for frontier AI companies to allow permanent third-party reviewers inside their organisations. These reviewers would have access to relevant tools and internal processes used to assess risks, according to Reuters.His broader argument is that AI companies should deliberately create more time between advances in model capabilities and the deployment of those systems, allowing researchers and outside evaluators to identify and address risks.Amodei also called for AI companies to work together voluntarily on common safety standards, as US lawmakers consider new regulations governing advanced AI systems.Story continues below this adWhy is Anthropic calling for a slowdown?The call comes as Anthropic has documented increasingly sophisticated attempts to use Claude for harmful purposes..Case attributions carry varying confidence, and Anthropic flags the weaker ones as such. Figures are the company’s own. Governments and companies named in the cases have denied the claims, said they were unaware of the report, or not responded.Source: Anthropic; ReutersExpress InfoGenIE · Global Desk.In its latest threat intelligence report, the company said it had identified and disrupted malicious operations between December 2025 and August 2026 across seven areas: cyber operations, influence operations, surveillance, scams and fraud, biological misuse, conventional weapons development and illicit model distillation.Anthropic said the activity showed that AI was moving beyond the traditional model of a person asking a chatbot questions. In several cases, threat actors used AI systems in multi-agent workflows that could conduct reconnaissance, exploitation and data collection with limited human supervision.The company said humans generally continued to make important decisions, such as choosing targets and reviewing stolen information. But AI could significantly increase the speed, scale and efficiency of these operations.(With inputs from agencies)