General AI
AI chatbots frequently recommend fake retail sites
The Verge AIhigh confidence
Why it matters
- →AI assistants are becoming primary shopping tools but lack robust verification against e-commerce fraud.
- →Perplexity's high error rate suggests current retrieval-augmented generation pipelines may prioritize relevance over authenticity.
- →Brands face reputational and financial damage as their official sites are displaced by impersonators in search results.
A study by IP protection firm Podqi tested three major AI chatbots—Gemini, Perplexity, and ChatGPT—using prompts related to 151 brands. The results showed that all three models frequently cited 'risky' URLs, defined as fake impersonator sites, dead domains, or knockoff retailers. Perplexity had the highest rate of risky links at 15%, followed by ChatGPT at 8%, while Gemini performed best with only 4%. The issue is particularly acute for regional brand queries and smaller brands, posing significant security risks to consumers using AI for holiday shopping research.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
How we got here
- 1Shopify's AI Search Boosts Traffic and Sales
TechCrunch AI · August 5, 2026 · Background
- 2Amazon AI Helps Identify Scam Messages
TechCrunch AI · September 2, 2026 · Related
- 3Amazon's Alexa Can Now Spot Fake Emails
The Verge AI · September 2, 2026 · Related
- 4AI Agents as Malware Distribution Channels
AI News · September 23, 2026 · Related
- 5OpenAI Dots Agent Review: Rough but Real
WIRED AI · October 7, 2026 · Background
- 6AI agents vs human web access
MIT Technology Review AI · October 19, 2026 · Background
- 7AI Browser Agents for Cross-Website Verification
Lev Selector · October 24, 2026 · Background
- 8ChatGPT launches verified Business Agents for e-commerce
The Verge AI · November 10, 2026 · Background
More from The Verge AI
© The Verge AIGeneral AIother
AI agent 'discovery' of Maya site debunked
A Google researcher’s claim that AI agents found a new Maya site collapsed when experts pointed out it was already documented in academic literature. The incident exposes a critical blind spot: autonomous systems lack the contextual knowledge to verify findings against existing research, leading to false positives and wasted expert time. While the agent successfully processed lidar data, it failed to cross-reference its output with known records, highlighting that automation without domain expertise is prone to error. This serves as a cautionary tale for using AI in scientific discovery, where verification requires more than just pattern recognition.
The Verge AI
© The Verge AIMarket & Regulationother
GDC Survey: 40% of Game Devs Use Generative AI
The GDC survey reveals a quiet but decisive shift in game development workflows. Forty percent of developers now use generative AI as part of their job, up from 36 percent last year. The adoption is heavily skewed toward free-to-play and mobile studios, where usage hits 58 and 70 percent respectively. While concept art remains niche at 22 percent, code generation and debugging have become standard tools for over half the industry. This data confirms AI has moved from experimental novelty to essential infrastructure in modern game production.
The Verge AI
© The Verge AIAgentsagents
ChatGPT launches verified Business Agents for e-commerce
OpenAI is turning ChatGPT into a storefront by letting brands deploy verified AI agents that handle comparison shopping and product inquiries. This moves the platform beyond passive information retrieval into active commercial transactions, with businesses controlling their agent's instructions via a new Business Hub. The feature separates these branded interactions from standard chats to maintain transparency, marking a significant shift toward AI-mediated commerce. It effectively positions ChatGPT as a discovery engine where brand identity and customer service converge directly within the chat interface.
The Verge AI
More in General AI
© WIRED AIGeneral AIother
Experts pivot AI safety from extinction to cyber and bio threats
The AI safety narrative has shifted from speculative extinction scenarios to tangible infrastructure risks. A WIRED panel of experts identifies nuclear integration errors, cyberattacks on critical systems, and biological weaponization as the immediate dangers. This marks a move away from abstract 'runaway intelligence' fears toward verifiable industrial hazards like flawed defense workflows and unregulated bio-threats. The consensus suggests that current lab-driven safety thresholds are insufficient for these concrete threats.
WIRED AI
© WIRED AIGeneral AIother
New 'Paper Trail' Tools for AI Authorship Proof
The Shy Girl scandal exposed a credibility crisis in publishing: authors have no standard way to prove they didn't use AI. A wave of startups like Workings and ObraVera are launching cryptographic 'paper trail' tools to timestamp drafts and fingerprint typing rhythms, aiming to create an immutable record of human creation. This shifts the burden of proof onto writers, forcing them to adopt surveillance-style metadata just to defend their voice against false accusations. While detectors remain flawed, these tools offer a technical shield in an industry paralyzed by distrust.
WIRED AI
© Lev SelectorGeneral AIagents
Website Prompt Injection Risks for AI Agents
A video demonstration highlights how malicious instructions hidden in web content can hijack AI agents through prompt injection attacks.
Lev Selector