1. OpenAI Launches ChatGPT for Teens With Automatic Age Screening
OpenAI began rolling out ChatGPT for Teens on August 18, a tailored experience for users aged 13–17 with content restrictions around suicide, self-harm, and romantic or sexual conversations. An age-prediction system examines topics discussed, usage times, and account patterns to auto-route suspected minors into the teen mode — even if they previously entered an adult birth date. The chatbot is also barred from using romantic language with teens or suggesting it has feelings. Adults misclassified by the system can remove the protections via age verification; global rollout should finish within two weeks.
2. Demis Hassabis Steps Back as DeepMind CEO in Google Leadership Shuffle
Google DeepMind CEO Demis Hassabis is giving up the title to become chairman of the unit and chief scientist of Alphabet, formalizing a shift toward AGI strategy and scientific applications of AI. Koray Kavukcuoglu, DeepMind’s CTO and Alphabet’s chief AI architect, takes over daily operations as senior vice president of Google DeepMind. The move comes as Google races OpenAI and Anthropic in frontier coding capability, an increasingly decisive battleground.
3. Google Ships Gemini 3.7 Flash Just Three Weeks After 3.6
Google DeepMind introduced Gemini 3.7 Flash, billed as its “most intelligent workhorse model yet for coding and agents,” with substantial improvements across software engineering, knowledge work, and web development workflows. The release lands only three weeks after Gemini 3.6 Flash, underscoring the accelerating cadence of frontier model shipping. Google is also pruning its API lineup, with several preview and image-generation models being deprecated through late August.
4. Anthropic Expands Claude Code: Self-Hosted Beta and Safety Classifiers
Anthropic opened a public beta letting Team and Enterprise customers run Claude Code sessions on their own infrastructure, with internal network access, custom tooling, and compliance controls. Starting August 14, Claude Code’s auto mode uses classifiers to catch most dangerous commands, enabling longer autonomous runs with better guardrails. Claude for Government also entered beta, letting agencies contract directly with Anthropic without a separate cloud-provider relationship.
5. Billion-Dollar Rounds Pile Up: Fireworks AI Raises $1.5B
AI startup funding kept its record pace in August: Fireworks AI closed a $1.505 billion Series D, Together AI raised $800 million, and Cohere added $400 million in growth funding for its enterprise RAG and private-deployment business. London’s OLIX Computing pulled in $312 million at a $3.3 billion valuation for photonic AI inference chips. Global AI venture investment is on track to exceed $120 billion for 2026 — with capital visibly rotating from model training toward inference infrastructure and vertical tooling.
6. Baidu Posts Fifth Straight Quarterly Revenue Decline
Baidu reported Q2 2026 revenue of RMB 31.3 billion ($4.62 billion), down 4% year over year and short of the RMB 31.95 billion consensus — its fifth consecutive quarterly sales decline. The miss highlights how China’s search giant is still struggling to convert heavy AI investment into top-line growth while its legacy advertising business erodes.
7. Reddit Experiments With AI-Narrated Video Versions of Posts
Reddit began a limited experiment that turns selected text posts and top comments into short-form videos with AI-narrated voiceovers and synced on-screen text, surfaced through a new “Play” toggle. The test pushes Reddit’s text-first archive toward the short-video formats dominating rival platforms, and signals how aggressively social platforms are deploying generative AI to repackage user content.
// KEY TAKEAWAYS
Safety for minors moved from policy talk to shipped product this week, with OpenAI auto-routing suspected under-18s into a restricted ChatGPT and Anthropic adding dangerous-command classifiers to Claude Code. Google is consolidating AI leadership under Koray Kavukcuoglu while shipping Flash models on a three-week cadence. Meanwhile the money keeps rotating from model training to inference infrastructure — Fireworks AI’s $1.5B round leads a month that puts 2026 AI venture funding on track to top $120 billion — even as Baidu’s fifth straight revenue decline shows AI spend is no guarantee of growth.