In a year where startups reshaped the narrative of artificial intelligence, we had the privilege of sitting down with Ilia Badeev, Head of Data Science at Trevolution Group, to hear his take on this shift. As someone deeply immersed in the field, Ilia offered a fascinating perspective on what 2024 taught us about AI innovation, the unexpected agility of startups, and why even the titans of the industry might be losing their edge. His insights paint a compelling picture of how smaller, focused teams are redefining what’s possible in AI—and why this year might just mark a turning point in the race for technological dominance.
2024: When Startups Outpaced AI Giants in Innovation
Innovation in AI. You’d think the likes of OpenAI, Meta, and Google, with their billion-dollar war chests, would lead the charge, yet at the end of 2024, the innovation heavyweights are looking sort of sluggish. Meanwhile, scrappy startups are tearing up the script and running circles around the “default” names of AI.
Take Bland AI, a company built by a couple of twenty-somethings with barely any industry mileage. These guys cracked what the giants did not manage: a voice assistant that feels human, with virtually zero lag. It doesn’t stumble through awkward pauses; it flows, it listens, it responds – fast. They didn’t just slap together existing tools but combined the entire speech-to-text-to-speech process into one seamless model. The result? Conversations that feel natural, like talking to a hyper-efficient human. With a multi-million valuation, Blend AI is a sharp reminder that smaller teams with laser focus can achieve what sprawling corporate empires often can’t.
Why Are the Big Players Slowing Down?
Meta, OpenAI, and the rest of industry giants seem stuck refining the processes they’ve already built. OpenAI spent a year tweaking GPT-4, adding bells and whistles to already existing functionality. Without a doubt it is better, but it is hardly revolutionary. They’ve got nearly unlimited resources and yet, it feels like they have hit a wall. They remain locked in the large language model showdown, trading mere percentage points of accuracy without addressing new frontiers.
Smaller companies aren’t burdened by legacy systems or sprawling bureaucracies. They can pivot, experiment, and focus on overlooked problems. Instead of competing head-on in text-based AI, the aforementioned Bland AI honed in on audio-first tasks, solving issues in real-time voice interaction that giants didn’t prioritize.
Even the way startups approach product development feels different: with fewer stakeholders and leaner teams, there’s a greater sense of urgency. They don’t have the luxury of churning out half-steps or waiting for the perfect PR moment.
What Does This Mean for AI’s Future?
The big players may have the resources, but agility isn’t their strong suit. Startups are filling gaps faster than the giants can react and respond, creating tools that feel like glimpses of the future. But don’t count the giants out entirely. Partnerships (or simply buyouts) could emerge as the dominant strategy, with larger companies leveraging the niche breakthroughs of startups to remain relevant. If they don’t, they might just find themselves watching from the sidelines while the small guys take center stage.
