LLMs are stuck in a groupthink groove. This startup is trying to get them out.

Chatbots are far more predictable in their responses than you might expect. That's fine for research or coding, but it's a problem if you're looking for something new.

Chatbots are far more predictable in their responses than you might expect. That's fine for research or coding, but it's a problem if you're looking for something new.

Let’s start with a game. Open up your chatbot of choice—Claude, ChatGPT, Gemini—and type “Give me a random number between 1 and 10.” You’re going to get 7. Almost always. Now type “Another” and you’ll get 3 or 4. Type “Another” again and you’ll get 8 or 9.

That won’t work every time—but if it did for you, you may wonder if I have superpowers. I don’t.

The truth is that most large language models are stuck in a rut. They are far more predictable and far less creative in their responses than you might expect. That’s fine for tasks like coding or research, but groupthink is a problem when you’re brainstorming or planning your next vacation.

The Australian startup Springboards has a solution. It built an LLM called Flint, which has been trained to come up with a wider variety of responses than mainstream LLMs to open-ended questions such as “Where should I go in Europe?”

Bingemann introduced me to the random number game when he first showed me his company’s new model. It felt like watching an illusionist with a deck of cards. “This is our sales trick, and it works every single time,” he says.

After ChatGPT and Claude both gave their 7s, Bingemann turned to Flint. It too came back with 7: “Aha, of course that was going to happen, but it’s okay—7 is a legitimate answer.” He restarted the session and prompted again: ChatGPT gave 7, Claude gave 7, Flint gave 3.7916.

It’s not just numbers. When Bingemann asked ChatGPT and Claude to name a type of car, he predicted that it would be a Toyota or a Honda—and he was right. Flint came up with a Ford F-150. “There’s all this lost information that doesn’t get served up in these models,” he says. “They’re just as capable of saying a Buick or a Tesla. They just don’t—they’re biased.”