For anyone who is currently circling the drain of AI existential dread, with AI companies claiming they've reached the singularity, and autonomous hacking, this is for you.
● In late 2025, October specifically, AI companies had scraped the internet dry of high quality books, conversation, and images, so to keep their billion-dollar valuations, they began training on AI output; they have managed to find fresh training data, though, not enough for long term success, and if the sources are correct, and they're training on synthetic data, too, they will begin to suffer model collapse; within three generations of synthetic training, AI systems forget human logic, and begin to spit out pure rubbish as they over-index on pure probability. They're so desperate, and going so far for training data because they know this. It's a matter of time.
● Meta, the company behind Facebook and Instagram, are threatened with a life-ending 1.4 trillion child endangerment suit.
● Anthropic's CEO has stated that AI companies need hundreds of billions in revenue to not fall into bankrupcy, with human workers continuing to be much cheaper.
● SpaceX (SCPX) plunged 20% in market value, and lost £600,000,000,000 in market capital in only two days; he isn't a trillionaire, anymore.
● Uber finds zero returns in replacing humans with AI.
● Sam Altman, CEO of OpenAI, who was harping on about how AI will take all jobs, rescinds that statement and says AI will not cause the job apocalypse he had prior predicted.
● Nvidia, the company behind AI ram & more, has stated, by an executive that "the cost of compute is far beyond the cost of employees."
● Experts state that xAI is a massive failure, and that the labs are risking a "big bubble explosion."
● Microsoft reports AI's cost problem, using the technology is more expensive than paying human employees.
● Thousands of CEO's admit AI had zero impact on employment or productivity.
● Starbucks abandons AI inventory tool after only nine months following multiple errors, stating, "they need to focus on consistency and execution at scale."
● Salesforce has scaled back use of AI model, citing, "reliability concerns."
● Sinch research reveals that 74% of enterprises have rolled back live AI customer communications agents.
● White-collar workers are rebelling against AI, as 80% outright refuse adoption mandates.
● AI giants, as a whole, are facing potential cost meltdowns.
● Corporate America is starting to ration AI, as the bill finally comes through, and costs skyrocket.
● Many AI researchers are sounding the alarm, as they leave the industry.
● BofA warns AI stock mania is nearing historic extremes.
● Ex-head of Google AI is warning that vibe coding doesn't belong in the workplace.
● Despite claims of AI coding being the future, many people who use it, are finding a net-loss in productivity, as it is crude, needing improvement.
● OpenAI's claim of automonous hacking comes one week after missing their financial goalpost by 90%, and losing another 20b; they have good reason to lie for hype.
● Claude Code has been removed from Microsoft engineers, as they find no improvements in cost or efficiency; this isn't a good sign for them, despite that they're also claiming three instances of automonous hacking.
It isn't all bad news, and they want you to believe it is, because they run on hype, and need you to believe it is the future, despite tech experts like Paul Allen, Jeff Hawkins, John Holland, Jaron Lainer, Steven Pinker, Theodore Modis, Gordon Moore, Roger Penrose, and more all claiming that trends suggest that technology doesn't achieve constant improvement, instead runs into decreasing returns, including AI.
In my opinion, these claims of reaching the singularity, automonous hacking are attention grabbing stunts, a distraction from their repulsive actions, which we must keep an eye on, rebel against and resist their attempts at monopolising information. They're showing their hand.
