Pangram’s AI‑detection tool claims near‑perfect accuracy, yet recent studies show it can be fooled by post‑processing techniques. The software’s global impact is felt as regulators and publishers seek reliable ways to spot machine‑generated content.
The Signal
Its vulnerability raises concerns about the future of content verification and the need for more robust safeguards against LLMs.