46
Why Europe has been absent from the great AI safety debate: shun the technology and lose out on growth; or embrace it and become dependent on tools developed by the US and China. (www.theguardian.com)
submitted 1 week ago by to c/technology@lemmy.world
36 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[–] 9 points 1 week ago (6 children)

And the EU has its own domestic AI, including Mistral which is one of the big ones. Not an AI fan myself but this is clearly nonsense

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 6 child comments
    •
  • [–] 3 points 1 week ago (4 children)

    And even if they didn't AI is rapidly becoming a generic commodity. Just build a datacenter and plug in an off-the-shelf LLM. There's nothing to really catch up on other than infrastructure, which will be radically cheaper when the bubble pops. There is no crucial or indispensible IP for countries or even companies to own.

    Given the current inflated costs (and the energy crisis), the best thing to do is wait until this shakes out and build in the future with a clearer understanding of the destination.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 4 child comments
    •
  • [–] 2 points 1 week ago (3 children)

    infrastructure, which will be radically cheaper when the bubble pops

    I wouldn't count on that unless the popping sound is nuclear. Demand for AI is not going anywhere, even when companies fail and investors lose out.

    What will improve prices for everyone is better algorithms and more efficient chips for them, helping supply to catch up with demand.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 3 child comments
    •
  • [–] 0 points 1 week ago (2 children)

    We must disagree on what a bubble is or whether AI is a bubble, because what you're saying seem nonsensical to me.

    Demand for AI will slowly grow, but the insane prices for components will come down because what is driving those prices is speculation about future growth, not actual current, near- or even medium-term demmand.

    Another thing to consider is that AI infrastructure is barely infrastructure at all--you have to buy new GPUs every 5-or-so years, at a tremendous cost. If you build a data center, they, in a sane world, those GPUS need to pay for themselves before the next batch of GPUs comes in.

    There really is no reason to go nuts building temporary infrastructure when the price is at its peak. Please, I'm begging you, start making sense.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 2 points 1 week ago (1 child)

    Current demand for AI seems anything but slow.. do you think this hardware is being hoarded by large buyers, instead of being put to commercial use ASAP?

    Whether it takes 3 months or 5 years to pay for itself is a valid Q though, and there is an element of risk that these companies take on with large hardware purchases in a rapidly evolving industry. Ultimately profits will need to cover any debts they take on, and there's only so much private investment money to go around... I expect to see more consolidation as this plays out.

    Owning datacenters with secure energy sources seems like a safe bet though, and is probably the bigger initial investment.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •
  • [–] 1 point 1 week ago* (last edited 1 week ago)

    Circular financing from NVIDIA is selling more cards than can be installed. The business press is absolutely supine on this, but it's likely there are qute a few purchased GPUs sitting in warehouses.

    https://infrathesis.substack.com/p/gpu-bought-cant-build-data-center-bottleneck

    Edit: "What do you mean people aren't finding a lot of gold? Don´t you see how many pans have been sold?!"

  • permalink
  • fedilink
  • source
  • parent