According to an article in WIRED, Google’s AI Overviews feature — the AI-generated summaries displayed prominently at the top of search results — is being actively exploited by scammers to inject fraudulent information into what users perceive as authoritative, Google-endorsed answers.
A WIRED investigation published February 15, 2026, confirmed that bad actors have successfully manipulated AI Overviews to surface fake customer service phone numbers, phishing links, and deceptive product recommendations.
This represents a significant escalation from earlier AI Overview issues (factual hallucinations, inaccuracies) to deliberate, targeted security threats that pose real financial and personal risk to billions of Google Search users.
How Are Scammers Exploiting AI Overviews?
The core vulnerability lies in how large language models source and synthesize information for AI Overviews. Unlike traditional search rankings that rely on established authority signals (backlinks, domain age, etc.), AI Overviews can be influenced by newer content that appears consistently across multiple low-quality websites.
Scammers exploit this by creating networks of sites that repeat the same fraudulent information, effectively manufacturing a false consensus that the AI model treats as reliable. This is fundamentally different from traditional SEO spam because it injects false data directly into what appears to be Google’s own authoritative summary.
What Should You Do To Protect Yourself?
- Never trust contact information from AI Overviews. Always navigate directly to a company’s official website to find phone numbers or support links.
- Cross-verify specific facts and figures from AI Overviews against multiple trusted sources before acting on them.
- Be skeptical of unfamiliar links surfaced in AI-generated summaries, especially those involving payments or personal data.
- Scroll past AI Overviews to traditional search results for high-stakes queries (banking, medical, legal), or consider using alternative search engines.
- Verify any phone number with a second search before calling it, as Google itself recommends.
Key Takeaways
- Google’s AI Overviews are being weaponized by scammers to surface fake phone numbers, phishing links, and fraudulent recommendations
- The attack exploits how LLMs synthesize information — creating fake consensus across low-quality sites to poison AI-generated answers
- This is a fundamentally new type of SEO attack that bypasses traditional search ranking authority signals
- Never trust contact information or links from AI-generated summaries — always verify through official channels
- High-stakes queries (banking, medical, legal) should bypass AI Overviews entirely and rely on known, trusted sources