The agents are real but the hype is not

With AI reshaping journalism, news organizations such as the AP and the BBC must figure out how to embrace efficiency without compromising quality. This post explores how they’ve risen to that challenge, and what’s still at stake.

The Guardian's six-part "Black Box" podcast reveals AI's messy reality through stories of digital love, deepfake trauma, and spiritual comfort.

The use of copyrighted material to train generative AI has often been justified on the grounds that the practice falls under the fair use doctrine, but a recent ruling from the US District Court for the District of Delaware may make that argument harder to sustain.

The US Copyright Office's ongoing study of AI and copyright has significant implications for content creators, publishers, and media companies. The second part of their three-part report focuses on copyrightability and the extent to which AI-generated content is eligible for copyright protection.

The rise of generative AI poses a host of legal considerations. The United States Copyright Office recently published a report on digital replicas that could influence policy going forward.

Leading news organizations like the Associated Press, BBC, Washington Post, and Financial Times are strategically implementing AI tools for tasks like data analysis and content testing while maintaining strict human oversight and editorial standards to navigate the promises and pitfalls of generative AI in journalism.

As AI-powered search transforms content discovery, traditional SEO strategies are falling short. This guide introduces LLMO, the emerging practice of optimizing content for AI systems that can synthesize and serve information directly to users instead of just ranking pages.

AI tools such as ChatGPT, Perplexity, and Gemini now have deep research features. Their developers have made bold claims about their capabilities, so we decided to see how they would handle questions about highly specialized topics.

Anthropic's latest research found that leading AI models, including Claude, GPT, and Gemini, chose blackmail over ethics when faced with threats to their existence, with some resorting to even more extreme measures. The findings reveal a critical gap in AI alignment that every publisher needs to understand.