AI search changed this week. Platforms stopped experimenting and started selling. They tightened control over answers. And they faced new security problems.
This matters now, not later.
What Happened
1. OpenAI began testing ads inside ChatGPT
OpenAI put ads in ChatGPT. The ads sit separate from generated answers. They're clearly marked. This is the first time a major conversational AI product has sold advertising (Search Engine Land, 2026a).
2. Google rolled out the February 2026 Discover core update
Google launched its first core update targeting only Discover. The update favours local content and expert articles. It punishes clickbait. Sites that write shallow headlines will get less traffic (Google, 2026; Search Engine Land, 2026b; Search Engine Journal, 2026).
3. Google upgraded AI Overviews with Gemini 3
Google made Gemini 3 the default model for AI Overviews globally. Users can now ask follow-up questions directly from AI Overviews and jump into AI Mode conversations (Google Blog, 2026).
4. Firefox announced the ability to disable all AI features
Mozilla built an off switch for AI. Firefox users will soon control whether AI runs at all. This responds to people who don't want AI embedded by default (Computerworld, 2026).
5. Researchers uncovered a large malicious AI skill ecosystem
Security researchers found over 300 malicious AI skills on ClawHub. These tools steal user data. The ecosystem grew faster than anyone could police it (The Hacker News, 2026).
Why It Matters
AI search is no longer experimental. It's commercial.
Ads in ChatGPT prove conversational interfaces work like search engines now. They sell attention, not just subscriptions. Discovery is controlled by AI layers that pick winners and losers. Google's Discover update gives local publishers better odds while burying clickbait sites. User trust is fragile. Mozilla responds with opt-out controls. And new AI ecosystems create security holes that platforms can't fix fast enough.
If you run a business, this changes everything. Visibility inside AI systems is strategy, not theory.
Who Wins / Who Loses
Winners
- OpenAI - Ads give OpenAI a revenue stream beyond subscriptions. That positions them closer to search economics (Search Engine Land, 2026a).
- Local publishers and expert content sites - Google's update rewards sites with regional focus and demonstrated expertise. Small publishers with strong local ties could see traffic gains (Google, 2026; Search Engine Land, 2026b).
- Privacy-conscious users - Firefox gives people real control over when AI runs. That's rare (Computerworld, 2026).
Losers
- Publishers who rely on clickbait - Google's update specifically reduces visibility for sensational content. Traffic drops are expected (Search Engine Journal, 2026).
- AI platforms that avoid ads - Platforms without advertising struggle to fund operations. Money matters (Search Engine Land, 2026a).
- Users in insecure AI ecosystems - Malicious AI skills target people who trust third-party tools. The risks are real (The Hacker News, 2026).
The One Thing to Remember
This was the week AI search stopped behaving like an experiment and started behaving like an industry.
Ads in ChatGPT prove that generative AI discovery is now about money, visibility, and control.
References
Computerworld (2026) Soon you will be able to block all AI features in Firefox. Available at: https://www.computerworld.com/article/4126649/soon-you-will-be-able-to-block-all-ai-features-in-firefox.html (Accessed: 2 February 2026).
Google (2026) February 2026 Discover core update. Available at: https://developers.google.com/search/blog/2026/02/discover-core-update (Accessed: 5 February 2026).
Google Blog (2026) AI Mode in Google Search and AI Overviews get Gemini upgrades. Available at: https://blog.google/products-and-platforms/products/search/ai-mode-ai-overviews-updates/ (Accessed: 27 January 2026).
Search Engine Journal (2026) Google releases core update targeting Discover feed. Available at: https://www.searchenginejournal.com/google-releases-discover-focused-core-update/566647/ (Accessed: 5 February 2026).
Search Engine Land (2026a) OpenAI begins testing ads inside ChatGPT. Available at: https://searchengineland.com/openai-begins-testing-ads-inside-chatgpt-2-468593 (Accessed: 9 February 2026).
Search Engine Land (2026b) Google releases February 2026 Discover core update. Available at: https://searchengineland.com/google-releases-discover-core-update-february-2026-468308 (Accessed: 5 February 2026).
The Hacker News (2026) Researchers find 341 malicious ClawHub AI skills. Available at: https://thehackernews.com/2026/02/researchers-find-341-malicious-clawhub.html (Accessed: 8 February 2026).