We’d like to remind Forumites to please avoid political debate on the Forum.
This is to keep it a safe and useful space for MoneySaving discussions. Threads that are – or become – political in nature may be removed in line with the Forum’s rules. Thank you for your understanding.
Doom scrolling and the next financial crash!
Comments
-
Major tech companies are investing heavily in AI technology and AI tech companies are using the data centres of major tech companies to power AI.
If AI falls the money invested falls, major tech companies suffer both losses in data centre demand and losses of investment and without demand from data centres hardware losses become greater.
In the last decade alone tech companies are the biggest dependency of the economy for everything including financial services.Monzo bank depends on major tech companies to survive.
Major tech companies looking at AI as a dependency on their products. AI companies are investing heavily to influence companies to use AI. If AI breaks it will affect everyone.
1 -
-
AI companies are doing their best to sell their produces, services and advertising. The initial excitement hasn't lived up to the hype and many companies now understanding that the promised efficiencies might not exist and if they do the cost of AI tokens is just too much. I expect we'll all end up using AI in some form, often without realizing it and certainly not by typing into ChatGPT and getting a dubious answer. But few of this first stage companies will stick around and trillions will be lost in this first bubble before AI is repriced at a sensible level.
And so we beat on, boats against the current, borne back ceaselessly into the past.2 -
An important risk is whether it will be the current US companies that will benefit when and if the supply of AI services becomes profitable. Deepseek from China is charging less than 10% of GPT's token costs, uses less computing resources and appears to match GPT for standard commercial activities. China is also benefiting from its massive investment in electricity generation over the past decades.
0 -
That 'trillions lost' doesn't just vanish though - it's being spent on something so owning whatever that something is would have been worthwhile.
0 -
AI will succeed. Any new entrant will need hundreds of $ billions of investment to catch up. China is cheaper, but will companies want to buy from them? Europe is the only potential new entrant with the necessary economic scale, I do not see any convincing signs of them stumping up the necessary investment. It is not as clear as the nay sayers would like us to believe.
0 -
It sounds dangerous to build your software and systems on an AI product that might be taken away at a moment's notice or might have some serious security issues. Already OpenAI is refusing to provide models to Cursor now that they are owned by Elon Musk's SpaceX. Embedding AI in our systems has two dangers; rouge AI actions and rogue actions by the humans who supply the AI.
And so we beat on, boats against the current, borne back ceaselessly into the past.2 -
If we are over building data centers then the excess will be scrapped because they are expensive to run and the GPUs will need replacing on a 3 to 4 year cycle. The intellectual knowledge will presumably be worth something and be used to rebuild, but the big losses will come from all the debt and dubious financing that defaults.
And so we beat on, boats against the current, borne back ceaselessly into the past.3 -
AIUI Deepseek is open source and open base data - it's all available for free. You can run limited jobs on your own PC and also modify the core data weightings locally to meet your needs. For basic AI tasks there is no need for a large central data centre.
ISTM that whether a US based supplier is less of a risk than a Chinese one could be open to question (Not here though).
1 -
Open source Chinese models or expensive proprietary US models? - Not much of a choice IMO.
And so we beat on, boats against the current, borne back ceaselessly into the past.1
Confirm your email address to Create Threads and Reply
Categories
- All Categories
- 355.6K Banking & Borrowing
- 254.8K Reduce Debt & Boost Income
- 456.1K Spending & Discounts
- 248.2K Work, Benefits & Business
- 605.7K Mortgages, Homes & Bills
- 179K Life & Family
- 263.5K Travel & Transport
- 1.5M Hobbies & Leisure
- 16.1K Discuss & Feedback
- 37.7K Read-Only Boards