Google DeepMind’s latest safety report indicates that, despite AI now writing over 75% of the lab’s production code, there has been no statistically significant acceleration in AI research progress since April 1, 2026. The capability index continues to rise at a modest 0.1 points per month, and while AI assists heavily in code generation, about 88% of code changes still require human oversight, with the Gemini 4 Argon model solving only half of its internal tasks, well below the 90% threshold that would signal full automation.
The findings temper expectations of rapid AI-driven productivity gains and suggest that immediate market enthusiasm for AI‑centric equities may be restrained. Investors may view the low risk assessment as a cue to maintain current exposure to tech stocks such as NVDA and INTC, while crypto assets tied to AI narratives, like ETH, are unlikely to see significant short‑term price movement. Overall sentiment remains cautious but not overtly negative, reflecting a measured outlook on AI’s incremental impact on capital flows.