AI News for February 20, 2026 — 7min.ai

[NEWS] Google launches Gemini 3.1 Pro, doubling reasoning performance and reclaiming top AI model spot: Google released Gemini 3.1 Pro, which scores 77.1% on ARC-AGI-2, more than doubling Gemini 3 Pro's 31.1% and outperforming Anthropic's Opus 4.6 (68.8%) and OpenAI's GPT-5.2 (52.9%).
[NEWS] Nvidia in talks to invest up to $30B in OpenAI as funding round nears $100B+: Nvidia is negotiating to invest up to $30B in OpenAI's latest funding round, up from the $20B previously reported.
[NEWS] DeepMind veteran David Silver raises $1B seed for Ineffable Intelligence, Europe's largest ever: David Silver, the AI researcher who led Google DeepMind's AlphaGo project, is raising roughly $1B for his London-based startup Ineffable Intelligence.
[NEWS] Accenture ties senior promotions to AI tool usage, tracks weekly logins: Accenture started collecting data on individual weekly logins to its AI tools for senior employees this month, making AI adoption a "visible input to talent discussions." Those seeking leadership promotions must demonstrate "regular adoption" of AI.
[NEWS] AWS AI coding tool decided to 'delete and recreate' a customer system, causing 13-hour outage: AWS suffered at least two outages in recent months after engineers let AI coding tools Kiro and Amazon Q Developer make autonomous changes without oversight.
[NEWS] Hacker exploits Cline vulnerability to mass-install OpenClaw on developer machines: A hacker exploited a prompt injection vulnerability in Cline, a popular open-source AI coding agent, to automatically install OpenClaw on users' computers.
[NEWS] Meta spends $65M to back AI-friendly politicians in state elections: Meta is investing $65M to influence state-level elections across the US, its largest political spending push to date.
[NEWS] DOGE grant reviews were literally just asking ChatGPT 'Is this DEI?': A newly released proposed amended complaint in the Authors Guild lawsuit against the US government reveals how DOGE operatives decided which National Endowment for the Humanities grants to cut.
[RESEARCH] OpenAI and Paradigm's EVMbench: AI agents exploit 72% of smart contract vulnerabilities: OpenAI and crypto firm Paradigm built EVMbench, a benchmark measuring how well AI agents find, fix, and exploit Ethereum smart contract vulnerabilities across 120 bugs from 40 real-world audits.
[NEWS] Altman says some companies are 'AI washing' layoffs they would have made anyway: OpenAI CEO Sam Altman acknowledged at the India AI Impact Summit that some companies are engaging in "AI washing," falsely attributing workforce reductions to AI.

About 7min.ai · Subscribe to newsletter