Perplexity under fireplace: Cloudflare says Perplexity’s AI bots faked their identities to bypass site blocks |

Perplexity, the rising AI-powered search startup, is going through scrutiny after Cloudflare alleged that its AI bots used misleading techniques to entry web sites that had explicitly blocked them. According to Cloudflare’s report, Perplexity’s bots masked their identification, altered person brokers to impersonate browsers like Google Chrome, and rotated IP addresses to evade detection. These actions allegedly allowed the bots to scrape content material from 1000’s of restricted domains. The accusations reignite considerations about AI firms bypassing consent protocols and exploiting writer content material with out permission.Cloudflare, one of many largest web infrastructure suppliers globally, claims this habits was noticed throughout tens of 1000’s of domains and tens of millions of each day requests. It additional argues that such stealth crawling undermines writer autonomy and violates transparency norms in net scraping. The controversy highlights rising tensions between AI companies in search of to construct data-hungry fashions and platforms or publishers making an attempt to shield their content material from unauthorized use.

Perplexity accused of evading restrictions by way of disguised bot behaviour

Cloudflare says it obtained quite a few complaints from purchasers who observed Perplexity’s AI crawlers accessing their web sites regardless of having put protections in place—similar to disallowing the bots in their robots.txt information or blocking them through Web Application Firewalls (WAFs). In a managed check, Cloudflare constructed new domains with restrictions in opposition to identified Perplexity person brokers. The firm claims that when the AI bots encountered these blocks, they re-identified themselves as regular customers by altering person brokers to mimic Chrome on macOS, and used rotating IP addresses that weren’t listed as Perplexity’s official infrastructure.

Perplexity responds, denies wrongdoing and blames confusion

Perplexity has pushed again strongly in opposition to the accusations, calling Cloudflare’s weblog submit a “publicity stunt” riddled with misunderstandings. An organization spokesperson said that Cloudflare failed to distinguish between Perplexity’s personal crawlers and third-party site visitors, particularly from companies like BrowserBase, which Perplexity says it makes use of solely sometimes. According to the corporate, the referenced 20–25 million each day requests have been largely user-driven and didn’t quantity to unauthorized scraping or bot-driven behaviour.

A rising flashpoint in AI and writer relations

Following the incident, Cloudflare eliminated Perplexity from its verified bot checklist and rolled out new measures to block its entry by default. The episode underscores bigger debates across the ethics of AI scraping, content material consent, and mental property. Cloudflare CEO Matthew Prince has warned that AI fashions pose an “existential threat” to publishers, and the corporate is actively encouraging web site homeowners to demand compensation for entry to their content material—signaling a brewing battle between digital infrastructure suppliers and AI companies within the period of large-scale content material consumption.