Italy’s Minerva-3B, trained from scratch on 2.5 trillion tokens, scores only 4.9% on Italian exams, raising questions about native-language investment needs.
Major AI labs publicly commit to automating AI R&D by 2026, signaling a strategic shift toward automation as a core objective, with broad implications for the industry.
Portugal’s €5.5M AMÁLIA LLM, launched in 2025, outperforms many models in Portuguese tasks but prompts key questions about openness, native data, and goals.
The Post-Labor Transition Atlas is a new empirical framework analyzing AI-driven labor displacement, policy responses, and structural alternatives as of 2026.
The Compounding Error Problem — Why 99.9% Alignment Decays to 60% in 500 Generations
Research shows that even 99.9% per-generation alignment accuracy drops to 60% after 500 generations, raising concerns over recursive self-improvement safety.
The Co-Founder’s Black Hole — A Structural Read on Jack Clark’s Automated AI R&D Essay
Jack Clark predicts over 60% chance that autonomous AI research systems could emerge by 2028, raising concerns about institutional readiness and future risks.
Jack Clark Says It Out Loud — Reading the Co-Founder’s 60%/2028 Estimate on Automated AI R&D
Anthropic’s co-founder Jack Clark publicly estimates over 60% probability that autonomous AI systems capable of building their own successors will emerge by 2028.
The Continual Learning Research Map: Where the Memento Constraint Stands in May 2026
Six months after initial analysis, the research community confirms the Memento Constraint remains a key bottleneck in AI continual learning, with no current solution ready.
Two Channels: How the Pentagon Just Split Frontier-AI Procurement in Half
The Pentagon announced a split in its AI procurement strategy, placing Anthropic in a separate cybersecurity channel and not in the classified, redundant channel announced May 1, 2026.
