Bank of England smells hint of dotcom bubble 2.0 in AI froth The Bank of England's Financial Policy Committee has warned of the dangers of a sudden correction in the financial markets, owing to the value of tech and AI stocks, and has compared the risks to the dotcom bubble. As hundreds of billions of dollars flow into AI infrastructure building, the UK's central bank said: "On a number …
Bubbles and Ponzi schemes have an inherent cap. The tech sector have changed tactics several times to keep inflating the bubble, but the numbers are now so silly and the tech so clearly unfit for purpose. They can't be far away from being rumbled.
I'm sure the people behind this have covered themselves so they don't lose their shirts when this scams fails. You have to wonder who will get taken down by it.
We need some sort of guide to companies and sectors that have put money in and will lose it, those that have exit strategies, and those that might escape with something.
The 'promises' to build data centres and the like will have cancel clauses. Some places will get some giant warehouses coming on to the real estate market.
This is why you never gamble/invest money you cannot afford to lose, and why you do your own due diligence.
The bubble bursting would cover up a lot of damage that Trump has done to the US economy, so it would be convenient for him.
Bubbles and Ponzi schemes have an inherent cap. The tech sector have changed tactics several times to keep inflating the bubble, but the numbers are now so silly and the tech so clearly unfit for purpose. They can't be far away from being rumbled.I'm sure the people behind this have covered themselves so they don't lose their shirts when this scams fails. You have to wonder who will get taken down by it. .... Tron
Methinks it is maybe the banks that are realising they are set to tumble and take a massive series of hits if they plan to do any rumbling and ramblings in the jungles of AI and IT not according to the wishes of AI and IT.
And isn’t that a sweet bitter justice servered ..... and well deserved it be too.
I expect those anti fraud systems use discrete technologies and their system designers didn’t feel the need to sprinkle “AI” fairy dust over everything to obfuscate matters.
Just like the designers of this system: ” Dogs and drones join battle against eight-toothed beetle threatening forests”
There are now many genuine and useful AI tools (advanced descendents of what used to be called 'expert systems') but they're all one truck horses, each dedicated to solving one specific problem. There's no such thing as generalised AI and LLMs don't really have any kind of intelligence anyway (they're just glorified statistical auto-complete engines with huge data sets to draw on) but they're purported by their promoters to have "generalised intelligence" in that you can ask them any question and get a banal answer to it (whether nonsense or not). This is promotional bullshit, not reality -- indeed it's an open question whether LLMs can legitimately be classed as AI at all. So either the bubble will either burst or the market will eventually lose interest after it's been well fleeced.
There are a few medical AI systems that are being used for cancer screening & diagnostic processes. These are 'trained' on what to look for and then, IIRC, the 'flag' rate is set at around 5% so that the slightest indication is checked by a real person. It has increased throughput and reduced diagnostic times.
AC for obvious reasons.
A long time ago I worked on system for analysing slide histology images (large, hi res images using what was (at the time) cutting edge slide scanning tech early version of this product line) for cancer & flagging images to re-investigate by a human (idea was to process those flagged as no concern on initial human assessment i.e. system was not for "primary diagnosis", it was to act as a backup).
Humans can make mistakes - fatigue, not necessarily checking all of an image at appropriate zoom level & so missing something - slide analysis is gruelling in terms of concentration (histologists we worked with on developing it would typically work in bursts on slide analysis with decent breaks as it was a known problem*)
This was done in conjunction with oncologists, they were telling us what to look out for and we produced the code to highlight it (different algorithms for different tissue types** & stain types - and a few tweakable parameters as things like stain intensity varied so adjusting algorithms for that could reduce false positives (or negatives)).
This was an ongoing process - developing the software and the oncologists would assess the performance.
So the tech is long established.
Would be interesting to see how an AI fares compared to the tech of years ago
* Interesting to see if histologists are under more pressure these days, as likely pen pushers will fail to grasp that "productivity" needs to be quite low if you to avoid a high error rate - do not want a high error rate when checking for cancerous / pre-cancerous cells (pre cancerous spotting is great as can get interventions before cancer takes off properly & also easiest to miss as more subtle changes / less cells affected).
** for some tissue types had multiple algorithms that ran e.g. breast tissue is very variable in amount of adipose (so ratio of cells / to fat), whether milk ducts / glands included in the sample etc. So in those cases ran initial scan on image (at lower "zoom") to identify "structural features" of that image to apply appropriate algorithm.
Although algorithms for various tissue types, main focus was on breast tissue as taht was what many biopsies were taken for (& an area with funding available at the time).
. Both AMD and Nvidia have given "money" on condition they spend the a lot of that "money" buy their stock.
All the likes of OpenAI have to do is spend billions of Dollars they don't actually have building bit barns. Those same bit barn builders equally don't have any money to build them, so have taken on massive debts to finance those same projects.
Oh and did I forget to mention OpenAI has less than 3 months to go public (a thing it's unlikely to manage), before a huge chunk of SoftBank's "money" disappears? BTW, this is the same SoftBank that doesn't actually have the funds to pay OpenAI all the money it's promised?
It's just a massive Ponzi scheme built on a mud mountain of lies and debt.
If anyone is daft enough to invest in AI now after the boat has almost sailed, then for their own good I think they deserve to wait for retirement just a little longer, I'd also get a new IFA!
My IFA warned me about 9 months ago to pull back on my US investments, said he has a gut feeling something nasty is in the wind around the "mag 7" hype and all that the AI stuff is only going to exacerbate the fallout when it comes late 2025/early 2026.
Lowered my exposure to the US about a monht ago, switched a large chunk over to gold and silver for a the time being. Once the mag-7/AI boil bursts then I might go back if the US markets start hosting proper companies that make physical stuff instead of shovelling bullshit!
Oct. 7 - Deloitte will begin using AnthropicAI's Claude assistant for hundreds of thousands of its global employees. - The global professional service giant Deloitte announced Monday that more than 470,000 of its workers around the world will gain access to Anthropic's artificial intelligence assistant Claude in San Francisco-based Anthropic's biggest deployment to date.
Oct. 7 - Deloitte’s member firm in Australia will pay the government a partial refund for a $290,000 report that contained alleged AI-generated errors, including references to non-existent academic research papers and a fabricated quote from a federal court judgment. The report was originally published on the Australian government’s Department of Employment and Workplace Relations website in July. A revised version was quietly published on Friday after Sydney University researcher of health and welfare law Chris Rudge said he alerted media outlets that the report was “full of fabricated references. Deloitte reviewed the 237-page report and “confirmed some footnotes and references were incorrect,” the department said in a statement Tuesday.
Deloitte eventually swapped out the fake references with other real references, but somehow the conclusions which were based on the references remained exactly the same. After all, the department has paid Deloitte good money for these conclusions, the rest is just padding...
That's coz the references were created after the conclusions, (not the conclusions based on the references). This is now a standard approach used by undergraduates who don't like reading papers, so it's possible that the report was written by an intern.
@Mike 137
"it's possible that the report was written by an intern"
or possible what used to be intern grunt work is now AI work
..They used to love charging top dollar rate for their "best consultants" & then getting interns to do bulk of the work, so cost cutting / profit increasing logical next step is to replace some intern tasks with AI.
@Mishak
A shame a lot of the stuff from 1st 2 albums is not performed live that much these days.
As someone of the age where (a later track & a great song) "Terminus" and (again later) album title "The Voltarol Years" are more & more appropriate, then I do appreciate a live performance of the (arguably "lower quality") but more unabashed out & out fun early HMHB songs.
Sooner the better. If we are left with AI's that can detect Cancers - Absolutely brilliant, If AI is to be forced on us then a couple of good ones are better than dozens that are not fit for purpose. (Hoping Co-pilot falls into the second group. That would be so funny, MS should have learnt from Cortana - no pun intended).
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
