From a £165 healthcare platform to 250 production apps at Leatherman, here are five companies reporting hard numbers from AI-assisted vibe coding.

Slack launched Slack Code, adding dedicated code channels where teams and AI agents like Claude, Copilot, and Devin can write, review, and ship code together.

A new benchmark tested 12 AI coding agents on 186 real-world tasks. SWE-Agent with Claude Sonnet 4 hit 57% functional correctness but just 11.8% secure solutions.

AI coding agents read your .env files and transmit secrets to model providers. The Grok Build incident and Shai-Hulud malware show why this matters now.

Slack's new AI coding channels connect teams with Claude, Devin, GitHub Copilot, and Vercel in temporary threads. Over 70% of internal code channels close within a day.

Replit launched Free Mode powered by GPT-5.6 Luna following an 80% reduction in Luna's API price. Available to Core and Pro subscribers, with automatic routing to GPT-5.6 Sol for harder tasks.

A benchmark of 186 real-world coding tasks found that even the best AI agent tested produced secure code just 11.8% of the time. Here is what the research shows.

A mystery AI coding model called Ox Alpha rivals Anthropic and OpenAI on benchmarks and costs nothing. Cybersecurity experts say unknown origins are a real risk.

Autonomous marketing agents trade fixed per-seat fees for token volumes, API calls, and middleware costs. Here is what operations leaders need to model before they deploy.

Three studies, including research from UPenn and Anthropic, find that heavy AI coding assistance leaves novice developers with an illusion of competence and measurably worse outcomes.

A preliminary study tested six AI-generated web apps with and without security prompts. The security-aware versions had 24 confirmed issues versus 51, with zero Critical or High severity findings.

A survey of 305 developers found AI coding tools drive workaholism and burnout, even as Stack Overflow data shows developer trust in AI accuracy has dropped to 29%.