AI Safety Worries Grow as Self-Improving Models Near Milestone
According to reports, US-based AI company Anthropic recently issued findings suggesting that fully autonomous AI development could be approaching sooner than previously expected.
The company stated that over 80% of the code used in its Claude chatbot is now generated by the model itself, a dramatic rise from single-digit levels observed in early 2025.
Based on Anthropic’s internal AI development framework, current systems are now assessed to have reached the fourth stage out of five. At this level, models are capable of executing code independently and assigning tasks to other systems without direct human direction.
The final stage in this framework is expected to involve fully autonomous AI agents that can design and train new models on their own, potentially enabling a cycle in which AI systems continuously generate improved successor versions.
According to reports, these advancements are unfolding amid ongoing warnings from experts, academics and industry leaders about the potential risks associated with increasingly powerful AI technologies.
Legal Disclaimer:
MENAFN provides the
information “as is” without warranty of any kind. We do not accept
any responsibility or liability for the accuracy, content, images,
videos, licenses, completeness, legality, or reliability of the information
contained in this article. If you have any complaints or copyright
issues related to this article, kindly contact the provider above.
Legal Disclaimer:
EIN Presswire provides this news content "as is" without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
