As AI systems become more prevalent in everyday life, a former CNN anchor-turned big tech executive is hoping to address concerns about bias in AI models, while also providing human expertise to complex or hot-button topics.
Campbell Brown, the former head of global media partnerships for Meta, and a former CNN primetime anchor and NBC Weekend Today co-host, has raised $3 million in seed funding for a new company called Forum AI, which will assess “how major AI models perform on complex and nuanced topics like politics, foreign affairs and mental health—areas where getting tone, balance and context is critical but where traditional data labeling falls short,” the company says.
Brown, along with her co-founder and fellow Meta veteran Robbie Goldfarb, have assembled a roster of experts and people with different perspectives, including well-known media figures like CNN anchor Fareed Zakaria, conservative CNN political commentator Scott Jennings, and journalist Salena Zito, as well as figures from the world of politics like Kevin McCarthy and Larry Summers, and entities like Mount Sinai Health System, the Manhattan Institute and the Atlantic Council.
“AI is already shaping how people understand the world,” said Brown. “We built Forum AI to make sure these systems reflect human expert judgment, not just statistical accuracy.”
Forum and its experts will evaluate AI outputs on sensitive or high-stakes topics, giving those companies “clear, independent feedback that helps companies improve reliability and trust.” And when there are gaps in knowledge, or there is breaking news, the experts can weigh in with key context to help fill those gaps.
Lerer Hippeau led the investment round, with Perplexity AI’s venture fund joining them.
Brown (who notes that she once hosted a show called No Bias. No Bull) says that Forum is hoping to bring a level of trust and transparency that is currently lacking.
“[F]or me, the real concern isn’t whether AI skews left or right, it’s seeing my teenagers use AI for everything from homework to news without ever questioning where the information comes from,” Brown wrote in a blog post. “Political bias misses the deeper issue: transparency. We rarely see which sources shaped an answer, and when links do appear, most people ignore them. An AI answer about the economy, healthcare, or politics, sounds authoritative. Even when sources are provided, they’re often just footnotes while the AI presents itself as the expert. Users trust the AI’s synthesis without engaging sources, whether the material came from a peer-reviewed study or a Reddit thread.”
