AI can give alerts / early warnings
the problem for humans is: how do we AVOID taking action on those alerts
I doubt humans really interested in solving real problems
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
-
Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.
Majority of CEOs Report Zero Payoff From AI Splurge
- Thread starter erek
- Start date
the problem for humans is: how do we AVOID taking action on those alerts
I doubt humans really interested in solving real problems
WhoBeDaPlaya
2[H]4U
- Joined
- Dec 16, 2002
- Messages
- 2,834
Me using LLMs at work
- Joined
- Aug 18, 2006
- Messages
- 11,472
My AI models jerk me off every time I ask a question. It's ridiculous. "That's such a pro dad move". "Wow, that's very smart"
Yeah no.
Twisted Kidney
Supreme [H]ardness
- Joined
- Mar 18, 2013
- Messages
- 4,721
The LLM's glazing isn't that surprising, the companies that own them want to soften the blow when they pull your pants down to check your tonsils.
- Joined
- Jul 13, 2006
- Messages
- 2,195
Huh...not mine....checks...yeah, definitely not mine...haha
How Elon Musk ended up with a $132-billion pay package
https://www.latimes.com/business/st...on-musk-ended-up-with-132-billion-pay-package
https://www.latimes.com/business/st...on-musk-ended-up-with-132-billion-pay-package
mullet
2[H]4U
- Joined
- Aug 19, 2004
- Messages
- 3,292
He earned every penny, good for him.
He didn't "earn" it ... he convinced a board of directors he was worth it ... and he hasn't even received it until he meets the targets, which is unlikely to happen because they were stupidly ambitious and they've even admitted that they're at risk.
An AI consultant tells Axios that one of their clients recently spent half a billion dollars in a single month after failing to put usage limits on Claude licenses for employees.
https://x.com/edzitron/status/2060006709557690841?s=20
https://x.com/edzitron/status/2060006709557690841?s=20
- Joined
- Jul 13, 2006
- Messages
- 2,195
Yeah, we won't know until the full accounting and reporting cycle for a company hits how much they intend to keep spending here.
It is incredibly ambitious, zero pay outside really high result that climb up to a giant payday (they are of course saying risk of getting $0 for a lot of work are high here):
would have well earned those bonus if he reach them, that said everyone was saying the same for the previous deal he had... and all the target got hit I think.
- Joined
- Jan 14, 2007
- Messages
- 17,023
Saw a clip of him agreeing with some firm that Tesla should be valued at 5 trillion... makes sense now. (he actually said 25 trillion later, but hey only 2 more years)
- Joined
- Jul 13, 2006
- Messages
- 2,195
https://www.forbes.com/sites/janaki...2026-ai-budget-in-four-months-on-claude-code/
Yep...the problem is these companies won't keep paying for something that expensive unless it can impact their bottom line by the capex +10% or more.
And for the numbers to work for the investments to pay the roic they need a lot of sustained income that has to translate into global gdp sized numbers. And the data is slowly coming in that it just isn't there.
Yep...the problem is these companies won't keep paying for something that expensive unless it can impact their bottom line by the capex +10% or more.
And for the numbers to work for the investments to pay the roic they need a lot of sustained income that has to translate into global gdp sized numbers. And the data is slowly coming in that it just isn't there.
If I recall correctly, the vast majority of his goals in 2018 were company valuation related which can, unfortunately these days, be done based on promises and hype. Plus, Tesla has stalled in revenue terms in the past 3 years. Musk went heavily political and pissed off a lot of his target customers (environmental types who want electric vehicles) so he's trying to pivot the company to robots. Robots would definitely be a safer revenue stream for a guy as much of a political hot potato as he is right now because those will primarily be sold to corporations, not individuals, and companies will only care about the bottom line and ditching workers for a robot that can work 24/7 without breaks. If that falls through I don't think car sales are going to do much for him right now. If he had stayed out of politics he'd probably be in a much stronger position to still be selling lots of vehicles and could still work on all the other projects. This is, of course, why the board has been demanding he stay the heck out of politics and focus on the company. Unfortunately, they can't wipe people's memories of the last few years and they can't muzzle the guy. If they could keep him off Twitter for a few years the whole lot of companies would be in a much stronger position.
A win is a win is a win. Nothing is ever guaranteed - still seems dumb not to bet even a little on that race horse, given their track record - which has upset decriers, multiple times on multiple occasions.
Panmure Liberum is the research provider behind this FT article. Their research estimates the five major hyperscalers (Amazon, Microsoft, Alphabet, Meta, Oracle) will spend approximately $658 billion in 2026 on data centres and related infrastructure.
Projections for hyperscaler + broader AI related capex are expected to exceed $1 trillion in 2027 or shortly after.
Broader market consensus (from various analysts) puts hyperscaler capex in 2026 in the $600B to $700B range, with some projections expecting the total to approach or exceed $1 trillion in 2027 as the build out continues.
Panmure Liberum argues that to justify these massive investments, the hyperscalers would need $2 trillion to $5 trillion in additional annual revenue by the end of the decade... an extremely aggressive growth assumption.
https://x.com/DelB0yTr0tter/status/2060181804565619167?s=20
Chinese game maker Mihoyo (Genshin Impact, Honkai Star Rail) accidentlly used ¥2 Million Yuan (USD $300k) worth of Token in 13 hours.
For large companies, even a single daily attempt without using "Tokenmaxxing" can be very costly.
On May 20th, Zheng Yinhe, the head of the AI NPC & Gameplay technology team of the "Honkai" series, inadvertently revealed the cost of an internal agent experiment at the 2026 Alibaba Cloud Summit:
An employee built dozens of agents to collaborate for a project, and as a result, it burned through tokens worth 2 million RMB (USD 300,000) overnight.
"We accept that there are costs and learning fees when exploring AI, and this also helps to improve our agent platform." However, some netizens pointed out that the value of these 2 million RMB doesn't seem to be significantly reflected.
Not every company can afford such "generosity". As some netizens joked, "A single 648 - yuan purchase is only a little over half a gold, and two million is just the money spent by three full - 65 hardcore players to support the game."
It's worth noting that on May 15th, Liu Wei, the co - founder of Mihoyo, said at an external event that the company will invest up to 100 billion yuan in the AI field in the next three years. He frankly stated that "even if it doesn't succeed in the end, we'll accept it. It's like setting off a big firework."
"As a company with the slogan 'all in ai', one day after spending 100,000 yuan on tokens, we blocked everyone's API," a netizen commented.
https://eu.36kr.com/en/p/3825876529238657
https://www.resetera.com/threads/chinese-game-maker-mihoyo-genshin-impact-honkai-star-rail-accidentlly-used-¥2-million-yuan-usd-300k-worth-of-token-in-13-hours.1534075/
Oracle down bad
Meta, Amazon, & Google feel the squeeze
(Google expected to survive because of TPU
Amazon might survive due to tie-ups with anthropic
Meta needs to pull their straps up)
Hyperscalers’ Free Cash Flow Dips as AI Arms Race Hits Balance Sheets
The full-year free cash flow at Amazon, Alphabet, Meta, and Microsoft is set to fall to its lowest level since 2014.
The decline reflects mounting pressure from heavy investments in artificial intelligence (AI).
https://finance.yahoo.com/sectors/t...perscalers-free-cash-flow-dips-083314165.html
Meta, Amazon, & Google feel the squeeze
(Google expected to survive because of TPU
Amazon might survive due to tie-ups with anthropic
Meta needs to pull their straps up)
Hyperscalers’ Free Cash Flow Dips as AI Arms Race Hits Balance Sheets
The full-year free cash flow at Amazon, Alphabet, Meta, and Microsoft is set to fall to its lowest level since 2014.
The decline reflects mounting pressure from heavy investments in artificial intelligence (AI).
AI Spending Spree Pulls Big Tech Cash Flow Down
According to recent estimates from Morgan Stanley, hyperscalers including Amazon, Alphabet, Meta, Microsoft, and Oracle could spend nearly $805 billion this year, up from an earlier projection of $765 billion. Forecasts for next year have also been raised sharply to $1.1 trillion.
https://finance.yahoo.com/sectors/t...perscalers-free-cash-flow-dips-083314165.html
- Joined
- Jul 13, 2006
- Messages
- 2,195
I'm shocked i say...shocked!!!
If only there was a way we have could seen this all coming?
(obviously sarcasm)
They all know this is going to happen, but it’s a competition and there will only be a few winners; the rest will be swallowed up by one of the winners...
google is now diluting its shareholding & divesting equity to public
tesla/spaceX/XAI is also going for ipo
openAI/Anthropic also probably eyeing IPOs
oracle is the most leveraged. will definitely need bailout
all the equity will probably go to pension funds
so the pension/retirement funds will end up taking a hit
Not a spoiler I don't think, but if anyone has seen Alien Earth they have a trillionaire tech CEO. Supposed to be dystopian and creepy but in a few years it will look as funny as Austin Powers "ONE MILLION DOLLARS!"
- Joined
- Jul 13, 2006
- Messages
- 2,195
View: https://youtube.com/shorts/njkU1gxSj4A?si=zQvCXHAsk_fThZlz
This is literally what's going to happen...fuck!
Some of my acquaintances (Europe) have been inquiring, and their funds aren't touching spaceX with a hundred foot pole.
The AI investment chickens have come home to roost.
By Bruno Ferreira published 5 hours ago
The move from subscription to usage-based billing has left many users looking down the barrel of massive bills, according to an Ars Technica report.
An ongoing discussion on the GitHub community forums includes plenty of customer testimonials, as do X posts from many different users. The overall gist is simple: many users are reporting that their bills would increase by several orders of magnitude, or that the limit is so low that a subscription plan is now either extremely limited or useless. There's even a community cost estimator that popped up a while back when the news first came to light.
To wit, the AI allowance in each of GitHub's subscription plans has three tiers: the $10 Pro plan gets you 1,500 credits; the $39 Pro+ plan contains 7,000; and the $100 Max subscription nets you 20,000 credits. While it's good for Microsoft to specify precisely how many tokens each plan includes, it's worth noting that in obvious cases like long-running conversations or queries on large projects, it's exceedingly hard to estimate how many tokens any given query will use, as shown by data from a 2025 paper.
Unsurprisingly, some users are reporting that even light usage, or being "super cautious", they went through significant chunks of their monthly allotment in the blink of an eye. Switching the underlying model for a query can drastically change the calculations; for example, using Claude Opus or GPT will be far pricier than using Gemini Flash. Some subscribers also warn fellow vibe-coders about resurfacing long-running conversations, as the nature of an AI bot requires the entire conversation to be re-sent again and again, quickly chewing through usage limits.
Besides being more judicious about selecting the right model for a given query, some folks are experimenting with harnesses to use token-efficient models like DeepSeek to keep costs down. Those using an AI bot as an agent also have to exercise extreme caution, as there's no shortage of reports and horror stories about clankers left unchecked, racking up gigantic bills.
https://www.tomshardware.com/tech-i...g-customers-report-up-to-100-fold-price-hikes
Github Copilot customers report up to 100-fold price hikes — AI sticker shock bites as Microsoft switches to usage-based pricing
NewsBy Bruno Ferreira published 5 hours ago
The move from subscription to usage-based billing has left many users looking down the barrel of massive bills, according to an Ars Technica report.
An ongoing discussion on the GitHub community forums includes plenty of customer testimonials, as do X posts from many different users. The overall gist is simple: many users are reporting that their bills would increase by several orders of magnitude, or that the limit is so low that a subscription plan is now either extremely limited or useless. There's even a community cost estimator that popped up a while back when the news first came to light.
To wit, the AI allowance in each of GitHub's subscription plans has three tiers: the $10 Pro plan gets you 1,500 credits; the $39 Pro+ plan contains 7,000; and the $100 Max subscription nets you 20,000 credits. While it's good for Microsoft to specify precisely how many tokens each plan includes, it's worth noting that in obvious cases like long-running conversations or queries on large projects, it's exceedingly hard to estimate how many tokens any given query will use, as shown by data from a 2025 paper.
Unsurprisingly, some users are reporting that even light usage, or being "super cautious", they went through significant chunks of their monthly allotment in the blink of an eye. Switching the underlying model for a query can drastically change the calculations; for example, using Claude Opus or GPT will be far pricier than using Gemini Flash. Some subscribers also warn fellow vibe-coders about resurfacing long-running conversations, as the nature of an AI bot requires the entire conversation to be re-sent again and again, quickly chewing through usage limits.
Besides being more judicious about selecting the right model for a given query, some folks are experimenting with harnesses to use token-efficient models like DeepSeek to keep costs down. Those using an AI bot as an agent also have to exercise extreme caution, as there's no shortage of reports and horror stories about clankers left unchecked, racking up gigantic bills.
https://www.tomshardware.com/tech-i...g-customers-report-up-to-100-fold-price-hikes
- Joined
- Jun 25, 2003
- Messages
- 28,297
AI companies have to recoup their investments somehow. And the reality sets in.
The AI industry's profit problem is getting bigger, says Goldman Sachs' equity research chief
It's still unclear whether AI can deliver returns that justify the enormous spending."All of the economic value has continued to accrue to the semiconductor companies," Covello said.
as companies continue to invest, the amount of profit needed to justify those expenditures is also increasing, he said.
That contrasts with previous technology cycles, when chipmakers prospered when their customers prospered.
Now, semiconductor companies have captured outsized profits, while companies above them in the AI supply chain have yet to show comparable economic returns, Covello said.
https://www.businessinsider.com/gol...ink-post&utm_medium=social&utm_source=twitter