Startup proposes crowdsourcing chatbots to deliver more reliable AI answers

John Davie wanted the Buyers Edge Platform, the hospitality procurement company he founded and still runs, to capitalise on the AI boom. After reviewing the options, the CEO felt none was quite good enough.

That led to the creation of CollectivIQ, a Boston-based startup incubated within the Buyers Edge Platform designed to provide users with more dependable AI query responses by displaying outputs from ChatGPT, Gemini, Claude, Grok, and up to 10 other models at once.

When the first wave of new AI tools started entering the market a few years ago, Davie said he was enthusiastic about their potential and encouraged employees across his company to begin experimenting with them. That excitement, however, did not last long.

“We had a bit of a wake-up call about a year ago when we learned that if our employees are just using any various AI tools, or even their own license, it could be training on our company information,” Davie said. “We could be essentially edging our competitor.”

Davie then explored enterprise-grade AI contracts that promised greater security. However, he said what he found were costly long-term agreements for large language models that still returned inaccurate information and frequent hallucinations.

“We hated having to decide which employees deserved AI,” he said. “What really made it worse, employees were complaining about hallucinated, biased answers. Sometimes it was really giving us flat, incorrect answers that made their way into PowerPoint presentations and cover presentations.”

He then pushed his chief technology officer to develop something better.

That effort resulted in CollectivIQ. The spinout developed software that simultaneously queries several major large language models, including systems from OpenAI, Anthropic, Google, and xAI. The platform then identifies shared information and points of disagreement across those outputs to create a combined response that is more accurate than any single model could generate on its own.

All prompt data handled through CollectivIQ is encrypted.

“As somebody who just loves technology, you’re always looking for the best of the best, right?” Davie said. “You always want to have the latest, greatest iPhone or laptop or tool, and I wanted to give my employees the best of the best of AI, but there was really nothing out there that you know would bring them all together into one.”

CollectivIQ began deploying the software internally for Buyers Edge Platform employees at the start of 2026. Davie said the early internal reception was very positive. After realising that many of Buyers Edge Platform’s customers faced the same uncertainty and hesitation about adopting AI tools, the company chose to make the product publicly available as well.

The platform was built on enterprise APIs from AI model providers. CollectivIQ covers the token costs itself, while customers are billed based on usage, a model Davie believes could help the company distinguish itself in an increasingly crowded enterprise AI space.

“I’m hoping that this is a breath of fresh air for companies that see that they are not going to have to be committed,” Davie said. “They’re only going to pay for the value they get out of it.”

CollectivIQ was entirely funded by Davie, who said he expects to look for outside investment at some point later this year. For him, the experience has also been personally energising, coming nearly 28 years after he launched his original company.

“It does feel like way back in the day, and we are doing it all over again and being scrappy and being very in the weeds on LLMs and post-training and all sorts of things I was not trained in,” Davie said. “It’s fun and exciting. I go sit hand in hand with the software developers building the product, that’s how I got my main company, it’s a lot of fun.”