The coolest thing about being a PR “hired gun” is that it requires you to learn the ins and outs of all kinds of different industries: sports, tech, financial, entertainment, etc. It’s absolutely mandatory when PR pros onboard new clients. (Which, if you stop and think about it, makes sense: you can’t optimize a company’s brand until you know how its profit model is supposed to work.)
Advertisement
In fact, I always warn people: Beware any marketing “expert” who tells you how to run your company before he bothers to learn your profit model! It’s almost always stupid, off-target advice.
The most common reason why most businesses fail is because they create products and campaigns for their own amusement instead of being laser-focused on the desires and aspirations of their customers. That’s also why most marketing efforts fail: The marketer is marketing to himself, not to his audience.
This brings us to the multibillion-dollar artificial intelligence (AI) industry.
Until just a few weeks ago, the conventional wisdom was that it was an investment-driven enterprise: the country with the most capital would upscale the fastest. The future, we were told, would belong to the first country that achieves AGI — artificial general intelligence.
And the two frontrunners are the United States of America (hooray!)… and communist China (boo!).
Supposedly, AGI is an inevitable landmark on the roadmap to ASI — artificial superintelligence. When ASI occurs, the reasoning and intelligence of our AI systems won’t just be on par with the world’s smartest humans; it will exponentially surpass it.
And before long, these AIs would become hi-tech gods, capable of calculations and breakthroughs that far exceed a million-trillion Albert Einsteins. The possibilities are near endless: cures for diseases, new solutions to complex social problems — who knows what our AI gods might do!
It’s a cool story. And it still might happen. But most people — and certainly most businesses — aren’t demanding an AI that’s a million-trillion times smarter than Albert Einstein. Instead, they simply want an AI that helps them finish a book report or summarize long, boring business emails or help their company with a handful of specific tasks.
Advertisement
Related: The AI Wars Begin: Stargate, MAGA, and the Technology to Dominate the World
Right now, it’s the biggest story in tech: China’s DeepSeek AI startup just released a chatbot model that reasons as well as ChatGPT’s most advanced (and most expensive) models. (Actually, some experts claim DeepSeek is better.) But just as importantly, it’s also cheaper: DeepSeek spent just $5.6 million to train its current model, compared to the hundreds of millions (or even billions) other AI companies spent. And it’s also significantly less expensive to run, with DeepSeek costing a mere $0.55 and $2.19 per million input and output tokens, compared to ChatGPT’s $15 and $60.
Of course, China isn’t exactly renowned for its strict adherence to truth, honesty, and playing by the international rules. It’s widely speculated that China might be fudging its numbers or acquired its training data via theft and/or corporate espionage:
“If you ask [DeepSeek] what model are you, it would say, ‘I’m ChatGPT,’ and the most likely reason for that is that the training data for DeepSeek was harvested from millions of chat interactions with ChatGPT that were just fed directly into DeepSeek’s training data,” said Gregory Allen, a former U.S. Defense Department official who now directs the Wadhwani AI Center at the Center for Strategic and International Studies.
Still, for most businesses, the bottom line is the bottom line. They’re not looking for an AI that’ll completely take over their company or be the next Albert Freaking Einstein; they’re looking for cost-effective tools to fulfill certain steps in their profit chain. If the company needs an AI to respond to emails or make sales calls, it simply needs an AI that’s good enough to do what it’s told.
Advertisement
They don’t want to hire Albert Einstein anyway. He’s overqualified.
Thus, for lots of businesses, governments, students, and organizations, an AI like DeepSeek is perfectly adequate for most jobs and tasks. You don’t need a $500 billion investment; you just need a “good enough” product.
Maybe AGI — and then ASI — are still on the horizon. And I’m sure the (highly-paid) researchers and (billion-dollar) executives at ChatGPT still believe in it. But at this point, the AI bigwigs are making products for themselves, not for their customers. They’ve lost sight of the marketplace.
And now they’re getting eaten alive by the Chinese.
