Anthropic has unveiled an updated version of its AI model Claude 3.5 Sonnet. According to the company, the model can now control desktop applications and perform PC tasks. It uses a new “Computer Use” feature, which is in public beta. Anthropic emphasizes that the technology is still error-prone and recommends developers initially test it only with low-risk tasks.
In addition to its enhanced capabilities, Claude 3.5 Sonnet is said to show improved performance in coding tasks. The company claims that the model outperforms competing systems in some benchmarks. Anthropic also announces the introduction of Claude 3.5 Haiku, a more cost-effective model with comparable performance to the previous top model Claude 3 Opus.
Experts see potential risks in the expanded capabilities of AI models. Anthropic states it has implemented safety measures to prevent misuse. These include classifiers designed to deter the model from actions deemed risky. The company emphasizes that it will monitor the use of the models and continuously adjust safety precautions.
Sources: TechCrunch, Anthropic