Are we being fed a steady diet of tech hype that’s more fantasy than reality? The truth might just shock you. From self-driving cars to AI chatbots, the promises often seem too good to be true—and they usually are. But who’s calling out these exaggerations? Enter Rodney Brooks, an Australian-born technologist on a mission to deflate the inflated claims surrounding world-changing technologies. And this is the part most people miss: Brooks isn’t a Luddite; he’s a pioneer. As a co-founder of iRobot (yes, the Roomba vacuum cleaner company) and RobustAI, he’s deeply embedded in the tech world. Yet, he’s also a vocal skeptic of the hype machine.
Brooks famously declared, ‘Having ideas is easy. Turning them into reality is hard. Turning them into being deployed at scale is even harder.’ In 2018, he published a bold list of predictions about technologies like self-driving cars, human space travel, and humanoid robots, vowing to revisit them annually for 32 years—until he’s 95. His predictions are categorized into three buckets: ‘Not in My Lifetime’ (NIML), ‘No Earlier Than’ (NET), and ‘By Some Date.’ But here’s where it gets controversial: In his latest update, Brooks admits he was ‘a little too optimistic,’ yet his realism remains razor-sharp.
Take self-driving cars, for example. Companies like Waymo claim their vehicles are ‘fully autonomous,’ but Brooks points out the glaring flaws. During a San Francisco blackout, Waymo’s robotaxis clogged intersections because they couldn’t handle dark traffic lights. Turns out, they often need human intervention—a far cry from true autonomy. Is ‘fully autonomous’ just a marketing gimmick? Brooks thinks so, and he’s not afraid to say it.
Humanoid robots? Brooks has been there, done that. In the 1990s, he built them at MIT, believing they’d seamlessly integrate into human life. Decades later, he admits they’re more theater than utility. ‘General-purpose robots that look and act like humans are chimerical,’ he writes. They lack human-like dexterity, stability, and safety. Ever wondered why Rosie the Robot from The Jetsons had wheels instead of legs? Animators in the 1960s got it right—a lesson today’s engineers seem to have forgotten.
Brooks also takes aim at large language models (LLMs), the tech behind chatbots. ‘They don’t answer questions directly,’ he explains. ‘They produce something that sounds like an answer.’ LLMs don’t understand facts; they predict the next word based on probability. Is this the future of AI, or just a glorified guessing game? Brooks argues we need ‘guardrails’—specialized LLMs for specific tasks, not one-size-fits-all solutions.
Here’s the bigger picture: We overestimate what new technologies can do and underestimate how long it takes to scale them. Brooks calls this out, reminding us that the hardest problems are always the last to be solved. Are we setting ourselves up for disappointment by buying into the hype?
So, what do you think? Is Brooks a much-needed voice of reason, or is he underestimating the pace of innovation? Let’s debate this in the comments—because the future of tech isn’t just about what’s possible, but what’s realistic.
