Pangram, an AI detection startup, has raised $9 million and partnered with Substack to identify AI-generated content. The company recently launched an AI image detection tool. Its CEO, Max Spero, joined TechCrunch’s Equity podcast to discuss the growing challenge of distinguishing AI-generated content from human-created material. The proliferation of AI-generated text and images is causing platforms and users to struggle with verifying authenticity, according to Spero.
Spero argues that determining how much AI was involved in creating content is more complex and potentially more valuable than simply labeling it as AI or human-generated. He also highlighted the risks of false positives, especially when they involve sensitive images. The stakes are high, as incorrect AI detection could lead to misinformation or privacy breaches. Spero further suggested that the 'bottom tier' of writing jobs may be permanently replaced by AI, while high-quality human writing could become more valuable.
The conversation on the Equity podcast explored the broader implications of AI on the internet, including the potential for a 'dead internet' scenario if trust in online content erodes. Spero emphasized the need for better detection tools and clearer guidelines to manage the increasing presence of AI in everyday content creation. The discussion also touched on the challenges of balancing innovation with accountability in an AI-driven world.
Source: techcrunch
