Modern venture capital has transformed into an unchecked oligarchy, or "Cancer Capital", where mega-firms managing $50B+ live off fees and shift risk elsewhere (Anil Dash)
ThinkingNews Desk · how this was written
Google released Gemini 3.8 Flash, its third Flash model in six weeks, offering a standard “workhorse” version for general tasks and a Cyber variant tuned for vulnerability detection. The model provides more reasoning steps and iterative tool calls than Gemini 3.7 Flash and is priced at $0.75 per million input tokens and $3.75 per million output tokens through year-end.
Written from all 4 reports below, not from any single one.
How it was reported
- TechMeme·Google launches Gemini 3.8 Flash, three weeks after 3.7 Flash launch, for an introductory price of $0.75/1M input and $3.75/1M output tokens until December 31 (Abner Li/9to5Google)
Google introduced Gemini 3.8 Flash three weeks after Gemini 3.7 Flash, pricing it at $0.75 per million input tokens and $3.75 per million output tokens through December 31. The new model is positioned as a next-generation intelligence for agentic workflows and cybersecurity, and Google claims it outperforms Claude Opus 5 and GPT-5.6 Sol on certain benchmarks.
- Hacker News·Gemini 3.8 Flash and 3.8 Flash Cyber
Gemini 3.8 Flash delivers significant coding and reasoning improvements over 3.7 Flash, outperforming larger frontier models on DeepSWE v1.1, Vals Finance Agent V2, Harvey’s Legal Agent Benchmark, and achieving 54.9% on HLE-Verified. It costs $0.75 per million input tokens and $3.75 per million output tokens. Gemini 3.8 Flash Cyber, available through the Fairwind Program, achieves frontier-level autonomous vulnerability discovery on CyberGym and a 70%+ success rate across 20 programming languages, while also supporting automated patching.
- TechMeme·Commerce Secretary Howard Lutnick says "we trust Anthropic" as they are "back on the right side" with the administration and that "they've done what we asked" (Maria Curi/Axios)
A federal judge ruled that Google must integrate its ad-tech tools with rival exchanges instead of selling its own ad exchange. Uber announced a 10% workforce reduction, cutting about 3,300 roles to focus on ridesharing, delivery and robotaxis. Google released Gemini 3.8 Flash, priced at $0.75/1 M input and $3.75/1 M output tokens, and the Trump administration filed a brief supporting OpenAI’s fair-use claim against the New York Times.
- Ars Technica·Google releases Gemini 3.8 Flash, its third Flash model in six weeks
Google released Gemini 3.8 Flash, its third Flash model in six weeks, offering a standard “workhorse” version for general tasks and a Cyber variant tuned for vulnerability detection. The model is available via API at an introductory rate of $0.75 per million input tokens and $3.75 per million output tokens until year-end, with regular pricing of $1.50 and $7.50. This release follows the 3.7 Flash launch and precedes the anticipated but unannounced Gemini 3.5 Pro.
- The Verge·Google says its new Gemini 3.8 Flash model ‘works harder’ but might cost more
Google released Gemini 3.8 Flash, which performs more reasoning steps and calls tools iteratively than Gemini 3.7 Flash. The new model keeps the same introductory price of $0.75 per million input tokens and $3.75 per million output tokens, but may consume more tokens for higher effort levels, potentially raising costs. Developers can continue using Gemini 3.7 Flash to reduce token usage.
- TechMeme·Modern venture capital has transformed into an unchecked oligarchy, or "Cancer Capital", where mega-firms managing $50B+ live off fees and shift risk elsewhere (Anil Dash)
Meta released Muse Spark 1.3, claiming a major boost in coding and agentic task performance at the same price as Spark 1.2, and announced upcoming “Watermelon” model and open-weight releases. Google launched Gemini 3.8 Flash, priced at $0.75 per million input tokens and $3.75 per million output tokens through year-end, touting superior benchmark results. Meanwhile, G20 members unanimously adopted the non-binding Carolina Principles, a US-proposed framework for lighter-touch, sector-specific AI regulation.