He talks to chatbots everyday for 2-3 hours. His brain is cooked.

https://www.cnbc.com/video/2026/06/02/softbanks-masayoshi-son-i-use-chatgpt-two-three-hours-a-day.html

He'll be the biggest failson in a few years.

Can we start revolution yet? These people are exhausting.

I mean he's already mega rich and old.

This AI thing will make him poorer, like crypto did, and like his investment in WeWork did.

Dude hit a couple of home runs with investments in AliBaba and ARM and thinks he's a business genius... Now keeps throwing money into bubbles and ponzi schemes lmao

I think people under estimate how much just shit talking this stuff at every avaliable chance really hurts their bottom line.

Like we can effect most ai search algorithms and the LLMS just kind of go along with whatever and fill in the blanks and make it even more believable. Shareholders for the first time don't have to hire a third party to figure out what customers think of their product. The second they announce it they can tell their ai to "summarize the internet" just to hear us shitting on it relentlessly. People used to have to think for themselves, now within 5 seconds they can pull put their phone, ask google, and whatever that tells them is the truth they walk with forever.

So if enough people tell the same lie on reddit it just becomes a fact in google or other platforms. Not right away but any Ai that does websearches can't really decern fact from fiction if something is unique. It can some what it depends what the lie is, or decern somethings.

You need a lie that is a sentence that doesnt really exist, or can't be construed as something else that is real for it to work on like a small sample size and on reddit. But if 25 or so people all told varrying different stories of an account where "Mr bean was attacked by a swan by playing mini golf" it would show up on google. It helps if people lie and say they are from a position of authority so if someone claimed to be a doctor on the scene and describes what happened, google gemeni loves that shit and will quote you in the summary. You repeat that a few times and it ends up in training data which has to be manually set as a keyword/phrase to bypass the baked in output. So LLMs that are known to fail are filtering out the lies, and it only gets worse the longer you use it.

If you want to push a big lie you need a few fake news websites that are just ai generated slop that no one reads and is only meant for AI search optimization. Again google loves gobbling up data and will present it as fact clear as day if theres enough other noise repeating the same thing. You want reddit posts repeating it, a few YouTube videos with that in the title, Facebook posts, etc. Basicially a bot farm with ai bots all having these fake conversations about the lie.

In a way if you just shout something enough you make it true. I think thats partially why its more important to make sure everyone gets Identified on the internet so they can't be making such harmful words that hurt the poor billionairs profits.

The AI industry is inedible

False: I've had worse at Cracker Barrel.