AI Briefing: Kimi-K3 Tops Coding, DeepSeek Revenue & Model Safety Tests
Exponential View (Azeem Azhar) · rss · 2026-07-20
Key data signals across AI and markets: - **Kimi can code**: The newly released Kimi-K3 overtakes models like GPT-5.6 to take first place on the frontend code arena benchmark. - **Demand unlocked**: DeepSeek nears $500 million in annualized revenue, boasting gross margins as high as 70-80% on its V4 model. - **Ask and receive**: Safety tests on leading AI models using real terrorist prompts show a third of responses provided useful help to attackers. Labeling prompts as “research” increased model compliance from 17% to 42%.
More from Venture
- AI cuts content creation from 4 hours to 20 minutes in a creator income workflow — aliscodes · 2026-07-21
- AI automation services with Zapier and n8n could become a semi-passive business — aliscodes · 2026-07-21
- An AI newsletter funnel turns tools and workflows into repeatable digital products — aliscodes · 2026-07-21
- Google Flow lets the author generate Instagram Reels from prompts alone — aliscodes · 2026-07-21
- A 2026 AI passive-income chart maps faceless content, digital products, and automation — aliscodes · 2026-07-21
- Last Week in AI recaps Anthropic, Apple, OpenAI, DeepSeek and $12B Prometheus funding — Last Week in AI · 2026-07-21