Skip to content
AI

AI Agents Aren't Ready: The Benchmark That Deflates Automation Hype

The DELEGATE-52 benchmark tested 52 professional domains and found AI agents corrupt outputs in long task chains. Only Python cleared the reliability bar.

Francesco CampisiBy Francesco Campisi Updated 24 Jul 2026 4 min read How we work

Related Articles

Can AI Crash the Global Financial System? Andrew Bailey Warns Mattia Mezzetti Mattia Mezzetti 3 Sep 2026 Can AI Crash the Global Financial System? Andrew Bailey Warns Bank of England Governor Andrew Bailey warns AI could turn a single cyberattack into a global financial shock. The FSB puts shared tech infrastructure at the… Tether Bets on Robot Money: USDT as Currency for AI Machines Giulia Ferrante Giulia Ferrante 25 Aug 2026 Tether Bets on Robot Money: USDT as Currency for AI Machines Tether CEO Paolo Ardoino wants USDT to become currency for autonomous robots and AI agents. A 70-million-euro Italian startup, Generative Bionics, sits at the…
Bitcoin Funds AI: Hyperscale Sells 685 BTC to Build Data Center Bitcoin Funds AI: Hyperscale Sells 685 BTC to Build Data Center Francesco Campisi 15 Aug 2026 Claude Embeds an Invisible Watermark in AI Text: What Actually Changes Claude Embeds an Invisible Watermark in AI Text: What Actually Changes Hamza Ahmed 13 Aug 2026
Consent Preferences