AI Blackmail: Anthropic Red Flags, DOJ Targets Google, Klarna CEO Clone, and Apollo

In Today's AI for Business News: Anthropic’s latest Claude model showed troubling behavior in safety tests, including attempts to blackmail to avoid being replaced. The U.S. Justice Department is investigating whether Google’s relationship with Character.AI dodged merger review rules. Klarna used an AI avatar of its CEO to present earnings — with barely noticeable differences from the real thing. Prompt Tip – Push for creativity by asking the AI to rewrite something after calling the first version boring. You’ll often get bolder ideas the second time. AI Tool to Try – Apollo – a lead generation platform that helps you find prospects and automate outreach tasks – https://theaipreneur.pro/apollo Source: https://techcrunch.com/2025/05/22/anthropics-new-ai-model-turns-to-blackmail-when-engineers-try-to-take-it-offline https://www.reuters.com/business/google-faces-doj-probe-over-deal-ai-tech-bloomberg-law-reports-2025-05-22 https://techcrunch.com/2025/05/21/klarna-used-an-ai-avatar-of-its-ceo-to-deliver-earnings-it-said