Google DeepMind co-founder Mustafa Suleyman warns AI is a ‘fundamentally labor replacing’ tool over the long term::Despite today’s AI hype, it’s still a “truly transformational” technology that will replace jobs unless policy steps in, Suleyman said.
IMO, the warning should be with companies, not individuals.
So, if a company needs fewer employees, there will be more companies, and more competition in the market. You lay off thousands of very smart people, and they’ll go to your competitors, or create new ones using the skills you gave them.
The problem right now is that companies are expecting AI to replace people right now. If you do, you’re in for a world of hurt as hallucinations and training issues will almost certainly give you some headaches - whether it’s a HR agent committing an offence that makes the company legally or financially liable, recruitment AI that rejects candidates disproportionately, or software tools that decide to invent features or introduce nested, poor-performing bullshit. This doesn’t even go into the liability issues of pumping your customer and employee data into third-party tools. AI in it’s current state is a useful tool, nothing more.
The competitors are also using AI. They don’t want to deal with “employees” any more than the big guys do. They’ll actually get more done in less time with AI. So at best, you stand out among the thousands that get laid off and some other employer lets you be the lucky one to gently steer the AI when it goes off course.
This is the most realistic outcome imo. Even before AI there’s already a pattern where employees jump through companies before they gather enough experience to make their own.
I think the next wave of work will be a huge bunch of mechanical turks to fight against hallucinations. These jobs will barely pay above minimum wage and you will be instructed to either tag data or generate new “pure human sourced” data.