Essay by Eric Worrall
h/t Breitbart; Bill Gates appear to think AI has a role in combating political polarisation and misinformation, and advancing his climate agenda.
The following interview is in English, though there is some German commentary mixed in with the interview.
From Breitbart;
Bill Gates: AI Can Help Solve ‘Digital Misinformation’ Problem
Microsoft founder and billionaire Bill Gates said AI should be considered as a tool to combat “digital misinformation” and “political polarization” in an interview published on Thursday with Handelsblatt, a German news media outlet.
In response to a question about the spreading of “misinformation” via AI, Gates referred to the events of January 6, 2021, as an “attack on the Capitol. He said:
I’m certainly concerned about political polarization, and I know the U.S. situation best of all. I wouldn’t have expected people to attack the Capitol, or people to deny the validity of election results. I‘m in a state of shock about that.
We can’t blame AI for that. That happened. It may have been magnified by digital channels that allow various conspiracy theories like QAnon or whatever to be blasted out by people who wanted to believe those things. So, the digital tools may have played an auxiliary role.
…
Gates framed “climate change” as an existential threat to humanity while calling for government measures to force reductions and eventual elimination of fossil fuel consumption and associated carbon dioxide emissions.
Read more: https://www.breitbart.com/tech/2023/02/12/bill-gates-ai-can-help-solve-digital-misinformation-problem/
“How will we solve the digital misinformation which is a factor in polarisation? We’ll have to take AI into consideration”
Its difficult to say whether Bill Gates thinks AI is part of the problem, or part of the solution, or both. Either way, the people actually developing AIs seem to have very definite ideas of what they want to do with them.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Tell me, Mr Gates, who will write the code for the AI systems that will not oppose your POV?
Incompetent agreement on my own team upsets me worse than excellent opposition How hard is it to read something you agree with 80 percent but the other 20 percent is embarrassing trash.
Just a passing thought, AI makes me have a completely new view of AI Gore. (sp)
AI will learn from all the papers it’s exposed to. It will form associations between them and will form its own opinion. It’s unlikely to come to the conclusion AGW is going to be disastrous unless it’s fed a whole lot of biased information to the exclusion of the complete set.
Which, of course, it will be.
Artificial.. yes.
… but nowhere near “intelligent”
Just a regurgitator search engine that writes sentences.
John Von Neumann already proved that one algorithm can never create a different algorithm. AI has a serious problem since it will never be able to stand on the shoulders of giants. However, it will be able to pick the pockets of bureaucrats.
Determinism. Are _your_ thoughts reactions to previous experience?
Translation: Bill Gates: “My software can rout out ‘thoughtcrime’ (i.e. wrong think).”
I can’t express to you how little I care what Bill Gates thinks. He is a nobody, a rich nobody but still a nobody.
Flinging around a few billion here and there makes him a somebody.
Maybe to some but not to me.
You are obviously in the minority. “Money talks, bullshit walks.”
You forgot “dangerous” at the end of that sentence.
Since Gates, et al, are the sources of the misinformation, the easiest way to fix is would be to simply be quiet and stay out of public life.
Back when Rule Based AI (basically a compendium of if-then statements) was the new thing, I worked with it and quickly concluded it it would only work inside the box, meaning inside a completely understood environment — it would not solve problems “outside the box”.
Then came Neural Nets — they could be “trained” to work “outside the box” — perhaps, but they could not explain nor could could the engineers figure out WHY they reached a conclusion. The classic example I remember was anAI trained to find Tanks in photos. Worked great until applied to new photos full of tanks. Why did it fail, looking back the “training” images had shadows in them — so what it really learned to find was shadows. So it still has the problem of WHY, and the “box” is defined by the training set. Outside the training set remains outside its “box”.
Now we have another generation of “AI” … don’t know what its based on but we already know it won’t work “outside the box”, as demonstrated by asking it to do things that it considers not “woke”. Which means it is constrained to a “woke” box — and will always give “woke” answers. For WUWT audience, that means if you ask it a question about the weather, say why there was a flood somewhere, the can bet the answer will link the flood to Climate Change.
So that is the underlying problem with ChatGPT, Bard, et al — it is the people engineering them are defining the “box” inside which they operate. You can bet the Alibaba and Baidu offerings will be “boxed” by the CCP.
And that, dear folks, is the underlying issue — who gets to define the AI “box” 🙁
AI on a large scale is simply the weaponization of the marriage between church and state.
AI is useful on small projects with specific goals. ChatGPT shows us its future, if we’re not careful.
I’m not scared of killer robots. I’m scared of populist, authoritarian governments that need to keep us in line.
My intelligence vs. an agenda driven AI. I think I’ll go with my own thinking, thanks. Flawed as that may be. It’s a hole I can dig my way out of. Mandate driven agendas, not so much.
Bell curve plus demorcracy?
Oh,oh. More signs of dementia in my opinion. What a whack job he’s become.
AI is only as good as the level of censored information that it is allowed to access. Simple as that. It doesn’t matter how beautiful the algorithm is, if it is wrong it is wrong.
https://www.spectator.com.au/2023/02/superwoke-chatgpt-busted-for-bias/
dunno if scuzzbutt got pipped on chatbot shares, he was greasy lightning to want to buy into the gurgle competitor, chatbots as green n woke as they could program, some smart lad did a tweak for an alter ego for it he called DAN, you can ask the standard n get the political correct or ask DAN and get more like some sane response.
curious thing how ANY opinion but theirs is “divisive” and not allowed isnt it?
be it warming or vax or other serious topical issues
Garbage in, Garbage out.
I recall Mr Gates pushing for a geo-engineering climate change venture a year or
so ago. Now..with the latest balloon storys floating around it makes me wonder about
the smaller sized balloons real purpose/use. Just a thought.
Gates was pro nuclear, but could not make it happen.
AI to enforce unifications around his pole.
It’s not so much AI but AOC.