The internet faces a growing trust challenge as AI-generated text and images increasingly appear not only in social feeds but also in job applications, product reviews and insurance claims. This spread forces platforms and users to reconsider what counts as "real" content and how to reliably identify AI-produced material.
Pangram’s role and recent developments
Pangram, a startup building AI-detection tools, recently raised $9 million to support its detection system. The company positions its technology as a "trust layer" for platforms, aiming to make it visible when text or images were created with the help of artificial intelligence.
Pangram has also entered into a partnership with Substack, which is using Pangram’s technology to indicate when newsletter authors rely on AI in their writing. In addition, Pangram has launched a new AI image-detection tool.
The expert debate: more than a true/false choice
Max Spero, Pangram’s co-founder and CEO, appeared on TechCrunch’s Equity podcast to discuss why AI detection cannot be reduced to a simple binary (real vs. fake) label. Spero argued that it is important to distinguish AI-assisted content from fully AI-generated work, and that detection systems should provide richer signals to users rather than a single yes/no verdict.
Why this matters
The spread of AI-generated content affects the credibility of social platforms, hiring processes and the reliability of consumer reviews. If platforms cannot reliably surface AI involvement, user trust and moderation efforts may be undermined over time.
Next questions
While Pangram’s funding and the Substack partnership represent concrete steps in addressing these issues, open questions remain: how accurately can AI be detected, how should disclosures be communicated to users, and where is the line between AI-assisted and AI-generated content? Max Spero’s discussion on the Equity podcast highlighted the complexity of these challenges without offering a simple, universal solution.
