Oscars Ban AI Talent

By the Dovetail editors · 4 min read

TL;DR

The Academy bars AI-generated performances and scripts from Oscar eligibility
GPT-5.5 matches Mythos Preview on cybersecurity benchmarks, suggesting broad model capability rather than a single breakthrough
Pentagon signs classified AI deals with major vendors but drops Anthropic over supply-chain concerns
Australia faces a policy inflection point on AI strategy

AI · Worldwide

Academy Awards now exclude AI-generated actors and scripts

The Academy has ruled that AI-generated performances and screenplays are ineligible for Oscars. The move draws a hard line between AI-assisted filmmaking tools and AI-generated creative output.

GPT-5.5 matches Mythos Preview in cybersecurity benchmarks

Independent researchers found OpenAI's GPT-5.5 performs on par with the heavily hyped Mythos Preview model in cybersecurity tasks. The results suggest advanced cyber capability is now a general property of frontier models, not unique to any one vendor.

Pentagon signs classified AI deals but drops Anthropic

The US Defense Department struck agreements with OpenAI, Google, Microsoft, AWS, Nvidia and xAI for classified network deployments. Anthropic was notably excluded after being declared a supply-chain risk.

Australian Tech

Australia's 'sliding door moment' on AI policy

InnovationAus argues Australia is at an economic inflection point comparable to the Hawke-Keating reform era. The piece calls for proactive national AI strategy rather than letting the moment pass by default.