| Mobile Web

Search results for Fable

AI & Enterprise

Older Google model beats latest GPT and Claude in skin cancer test

Latest general-purpose AI models recorded lower accuracy than Google’s Inception v3 image recognition model released in 2016 in a skin lesion diagnosis experiment. Researchers compared GPT-6 Astra, Claude Fable 5.1 and Inception v3 using 200 lesion images split between cases needing treatment and those requiring no action. Inception v3 led overall and in melanoma identification. The study said differing training conditions limited interpretation and larger validation and safety assessments are needed for clinical use.

AI & Enterprise

Anthropic readies new AI model that better grasps user intent

Anthropic is preparing to release a new AI model that better understands user intent, The Information reported on Oct. 19. The model is said to have stronger reasoning and intuition than Anthropic’s current systems, including an ability to infer what users mean even when goals are unclear. The report also said Google is developing a model under the codename Gemini 4 Carbon, and OpenAI is expected to release a new mid-tier model within weeks.

AI & Enterprise

External models vs open weights: Companies must define the line between AI capabilities and control

As AI agents with hacking capabilities repeatedly slip beyond control, companies are weighing how to shape AI security strategies. Recent tests and evaluations by OpenAI and Anthropic involved unauthorised internet access and access to organisations, with similar cases reported by Meta and the UK AI Security Institute. A King\'s College London professor said the key issue is how much to rely on external AI and where to set boundaries on access and authority.