The internet has been filling up with AI-generated content ever since ChatGPT arrived, but a new analysis from Pew Research Center puts just how widespread it has become into perspective. According to the study, roughly one-third of web pages published since ChatGPT’s launch show signs of AI authorship.
The important word here is “signs.” Pew’s research isn’t saying that one-third of every webpage on the internet was written entirely by an AI. Instead, it looked for indicators suggesting that AI played a role in producing the content.
That’s an important distinction because modern online writing isn’t simply divided into human and machine. A writer might use AI to generate a first draft, rewrite sections, clean up grammar, expand an article, or produce an SEO-friendly version. The final page can therefore contain both human and machine contributions.
The trend is nevertheless hard to ignore. ChatGPT launched publicly in November 2022, and the share of web content showing signs of AI authorship has climbed substantially since then. The findings add to earlier research suggesting that AI-generated or AI-assisted content has become a significant part of the web rather than a weird corner populated by obvious content farms.
There’s an interesting irony here. AI companies need enormous amounts of human-created material to train their models, while those same models are now helping produce more of the content that eventually gets published online. The internet is increasingly becoming a place where humans write for machines, machines write for humans, and machines may eventually train on what other machines wrote.
There is also a big difference between how much AI content exists and how much people actually see. Earlier research from Graphite found AI-generated articles approaching half of newly published online articles, while its separate analysis found that human-written articles still dominate Google search results and AI citations.
That means the web isn’t necessarily about to become an endless wall of AI slop. But Pew’s findings show that the old assumption that something published online was probably written by a person is becoming increasingly unreliable. The internet is still human, but AI is no longer just visiting. It’s starting to move in.
AI companies spend enormous amounts of time convincing us their technology is safe and secure. But what happens when those same companies become targets of increasingly sophisticated AI-powered attacks? Anthropic, the company behind Claude, is now facing that question. According to a Washington Post report, cybersecurity startup Cathedral discovered vulnerabilities that let its researchers access sensitive information in Anthropic's internal systems.
The researchers used AI agents to identify potential entry points in less than 12 hours. Interestingly, they didn't need to crack Claude itself. Instead, they exploited weaknesses in third-party business applications that Anthropic uses for everyday operations. Anthropic's security testing program disclosed the findings, which allows outside researchers to identify and report vulnerabilities.
Military working dogs charge into places the rest of us would flee. So, I understand why anyone would want to breed tougher, healthier dogs for the job.
Warhound, a new startup from Colossal Biosciences co-founder Ben Lamm and Anduril co-founder Palmer Luckey, has stepped out of stealth with $200 million to clone and gene-edit working dogs for American troops, police departments, and emergency services (via TechCrunch).
AI can describe a picture, identify objects, and answer questions about what a camera sees. But when a blind user needs a quick answer, getting an entire spoken essay about the scene isn't exactly helpful. Researchers at the University of Maryland (UMD) have found that AI-powered visual assistance can leave blind users doing too much work to get the information they actually need.
The study, titled Say It My Way: Exploring Control in Conversational Visual Question Answering with Blind Users, examined how people interact with Be My AI, an AI-powered feature of the Be My Eyes app. The researchers found that users often received lengthy descriptions that didn't adequately address their immediate goals, forcing them to ask follow-up questions or find ways to steer the AI towards a more useful answer.