Legal professionals in South Africa have been warned that AI-generated case law research must be independently verified by lawyers due to the risk of AI hallucinations compromising case integrity. The warning emphasizes that human judgment must remain central to judicial and legal processes rather than being replaced by AI systems.
Source: ITWeb
The article argues that public backlash against AI data centers in the US is primarily driven by concerns about electricity consumption and grid strain, rather than abstract anxieties about corporate power as tech industry leaders claim. AI data centers require enormous power capacity—ranging from 300 megawatts for smaller facilities
Anthropic has released new models called Mythos and Fable that demonstrate improvements in coding, reasoning, and vision capabilities. The company reports these models communicate more clearly with users. Source: TechCrunch
Reports indicate Jensen Huang, Nvidia's CEO, had an alleged affair with a person suspected of espionage, which may affect Nvidia's efforts to influence U.S. export policy on AI chips to China. The situation could complicate the company's regulatory and diplomatic objectives regarding access
Meta has released Muse Ember, a lightweight language model designed for agent execution priced at 10 cents per million input tokens. The model is positioned as a cost-competitive alternative to existing lightweight models from other providers. Source: TechCrunch