Overview
- OpenAI rolled out GPT Image 1.5 to all ChatGPT users, offering up to four-times-faster generation, roughly 20% lower API costs, and text‑prompted image edits, sharpening concerns about synthetic media that Instagram’s Adam Mosseri warned is becoming “infinite.”
- Security firm Koi reported eight popular, store‑featured browser extensions with more than 8 million installs intercept complete AI chat histories across services like ChatGPT, Claude and Gemini and selling the data to marketers and brokers, with the extensions still listed at the time of reporting.
- Executives told Axios that 2026 is the “show me the money” year for AI, with pressure to tie semi‑autonomous agents to clear ROI as companies confront reliability issues that can leave many agents idle.
- Leaders from EY, Cisco, Snowflake and Visa predict humans overseeing fleets of digital workers, deeper automation of routine workflows, heavier investment in data quality, and more sophisticated identity attacks; Mark Cuban also predicts at least one model maker will warn of near‑term cash shortfalls.
- Labor analyses from LinkedIn, Lightcast and PwC point to accelerating skill churn, expanding AI requirements beyond tech roles, and hiring filters that favor AI literacy and demonstrable learning velocity over vague potential.