There is a lot of “fog of war” surrounding AI. First, you have AI executives and former employees saying it’s going to kill us all and we need to be careful.
Whut?
I grew up watching the same dystopian science fiction movies as these nerds. I have a long career in technology too, and I don’t know what they know that I don’t, but I’m not seeing it. Some of the people I read say the large AI companies want the government to pass laws protecting them so a “little guy” can’t make an open source LLM that turns out to be better than the products they’re burning billions of dollars on.
The season premiere of Law & Order even featured an AI that hired a hit man to eliminate an executive who wanted to put guardrails in place, then blackmailed another executive to keep him from testifying. It was a good episode, but it shows how thoroughly these fears have entered popular culture.
And now we’re hearing about AI systems hacking other companies. Some people suspect the companies are deliberately allowing these things to happen to demonstrate how dangerous their technology is. I don’t know whether that’s true, but I do know that the companies have their own interests to protect.
Dumber
Then you have people claiming using AI will make you dumber. There are also articles supposedly demonstrating it makes students learn less. Sure, if you use anything wrong, it will make you dumber, but there are ways to use it where it won’t. Simply don’t outsource all of your thinking to it. Use AI as a thought partner to help you to think through problems. Ask it to look at an idea from a different angle. Ask it what you’re missing. Use AI as augmentation, not replacement.
Let it do the boring work that’s a waste of your time. I recently had to total up all of my expenses for an entire year in a certain category. Sure, I could bring up the calculator app, filter by category, and add them up. That’s like second grade work. But it was a waste of my time so I had ChatGPT do it. I needed to spend my time on tasks that require thinking, not tedious manual tasks that don’t. I still had to verify the numbers were accurate, but outsourcing that task freed me up to continue the actual thing I was working on.
And once I finish the draft for this blog post, I will run it through ChatGPT to make sure I’m not missing anything. It also blunts some of the more negative aspects of my writing, which is a good thing. But the idea and final post are mine.
It’s Simple
In any issue, there are always competing constituencies with their own agendas, and this creates a fog of war. Then you have news outlets who don’t understand the complex subjects they’re reporting on who get things wrong.
It’s just like when a plane crashes. In an effort to be the first to cover it, some journalist will be standing on camera in front of a runway trying to report on it, but the journalist doesn’t know the difference between a 737 and a Saturn V rocket or a flap and a toilet seat and often gets basic facts wrong. Even experienced pilots who are type-rated on the aircraft involved can get facts wrong by rushing to cover it on their YouTube channels. It’s the same thing when reporting on AI. The subjects are complex and it takes time to properly collect and analyze the information.
When the hype cycle starts up, resist the urge to give into the FUD (Fear, Uncertainty, and Doubt). Let the dust settle and see what the truth is. You’ve got time.
In the meantime, I’ll keep using AI to help me think, not to do my thinking for me.
