Anthropic's IPO, and those that will follow

 

Anthropic is trying to raise billions of dollars in its Initial Public Offering of shares, to fund its projected growth in the artificial intelligence (AI) market.  Karl Denninger retorts:

After hyping every single possible use for "AI" and firing people en-masse to replace them with computer "intelligence" firms are discovering that the code and processes produced don't provide an actual economic benefit all-in compared with skilled employees.  Firing people of course saves money but then the compute you buy directly and indirectly costs money and what the machine produces had better not require much "spit and polish" or you're underwater.

Worse, and I've discussed this before, doing this makes the technical debt percentage on said generated stuff nearly 100%.  That is, you have no humans that understand it at all because no human built it and if a human reviews it that is akin to reading a book as opposed to writing it.  The reader might (in this case will) skim it and lacks deep understanding -- specifically they have no understanding at all why certain decisions were made when it comes to structure because no human made the decision.  The alternatives were weighed (if at all) in a machine rather than a human and the machine may or may not retain enough context to help you in the future.  Five or ten years hence it will almost-certainly be "not" in that regard.

Firms have already run into this and many who have realize the problem -- as this accelerates it becomes a deeper and deeper hole.  When (not if) something goes wrong there is nobody who can quickly look at the code and figure out why it broke.

There's more at the link, all of it very ugly from Anthropic's point of view.

Lawrence Person provides a different, but still damning perspective.

Anthropic made just under $1 billion in profit for one quarter (it’s first profitable quarter), yet it has committed to spending an additional $1 trillion, so it only needs 250 years of similar profits or so to pay off its projected capital costs!

. . .

If a small number of giant tech firms are just trading IOUs back and forth, how much real revenue is actually being generated by any of them? It’s less a self-licking ice cream cone than a self-climbing rope untied at the top, with each of the tech giants acting the fakir in turn.

Then there’s the big risk of the whole AI smoke and mirrors edifice collapsing: “If revenue growth stalls and fresh capital dries up, the company could breach covenants under its credit facility and commercial agreements, triggering cross-defaults and accelerated debt repayment.”

AI has proven that you can actually do useful things with it, and Anthropic’s Claude was (the last time I checked) the leader in AI code generators. But the amount of revenue coming into AI companies is nowhere near the trillions of dollars of capex being shoved into the insatiable maw of the AI beast.

Again, more at the link.

Anthropic is, of course, only one of half a dozen or so giant AI corporations and conglomerates vying to dominate the market.  Of them all, I think only two or three, including SpaceX, Google and possibly Apple, generate (or are likely to generate) enough revenue from their existing resources to cover the costs involved - yet most of the others are already committed to borrowing enormous sums of money that they have no realistic chance of repaying out of earnings.

What happens when the train wrecks?

It's going to be a wild ride . . .

Peter

Assassination and sabotage just became much easier to accomplish...

 

I was intrigued to read about quadcopter drones being used to carry short-range missiles that could be fired against incoming drones.

A modified version of Malloy Aeronautics’ T-150 Uncrewed Aerial System (UAS), designated TRV-150, intercepted multiple one-way attack drones during trials at Eglin Air Force Base, Florida, in late September, according to BAE Systems. The U.K.-developed platform fired laser-guided Advanced Precision Kill Weapon System (APKWS) rockets from a three-pod launcher, engaging successive targets without landing to reload.

The test took place as part of a Mission Capabilities Joint Readiness Experimentation exercise  of the Department of War’s (DoW) Office of the Under Secretary for Research & Engineering, or OUSW(R&E).

The latest test builds on an earlier APKWS firing demonstration at the Dugway Proving Ground in Utah in 2025. On that occasion, the drone conducted single-shot engagements against both air and land targets, while the new trial introduced a three-pod launcher designed to engage multiple aerial targets in succession.

There's more at the link.

I'm sure the drones and their missiles will add another very useful layer of protection against incoming enemy drones.  However, there's another aspect to this that worries me.

These drones are small enough to get close to an intended target without being detected by most current defensive systems.  A quadcopter drone can fly down a narrow road, or along a firebreak in the forest, so low that it can't even be seen.  It can literally fly behind something like a delivery van if desired, and the crew of the van would never know it was there.  As soon as it got within range of the target, it could pop up over surrounding terrain, launch its missiles (and, if need be, designate their target with a laser beam), then duck down and make its escape the same way it arrived.

Think of the targets that could now become much easier to strike, because nobody would know the threat was approaching in time to do anything about it:

• A refinery, processing plant, etc.;
• An armory or aircraft on a military airfield;
• Oil tanks or vital equipment in a harbor;

and so on.  The list is almost endless.  In fact, one of the most worrying aspects will be the physical security of our leaders, both political and military.  If a national figure is scheduled to deliver an important address or be present at an important function, or a senior military commander is to visit units in the field, what if several such drones could approach unseen and salvo their missiles at him without warning?  His chances of survival would be minimal at best.

Terrorist groups and foreign powers will have no trouble at all getting their hands on such equipment.  (In fact, the guidance system used by APKWS has already been fitted to non-US rockets with complete success.  How many other nations have access to such rockets, even factories to produce them?  The number must be in the dozens or scores by now.  Even if compatible rockets are not available, there are competing rocket systems from several countries that can do the same job, and can presumably be launched in a similar manner.)

More and more, technology is a two-edged sword.  We can't do without it, that's for sure, but how long can we live with its proliferation?

Peter

Fighting the crud, and heightened flood danger

 

I got nailed Friday by the latest wave of flu sweeping the country, and spent most of the weekend feeling like death warmed up.  It hasn't improved much, so I haven't had the time or mental capacity to sort through likely memes for Monday.  I'll do my best to have an improved crop next week.

However, I do want to make an urgent suggestion for those in the path of the increased winter precipitation forecast for the Super El Nino scheduled to hit us over the next few months.  I did some research.  It looks as if rainfall in large parts of the South will be much heavier than usual, possibly causing flooding in areas that have not previously been at risk for it.  Add to that danger the rapid expansion of suburban construction:  for example, we have a new subdivision going up within a mile of us, and uphill from us at that.  All the rain that falls on its roofs, concrete, roads, etc. will no longer have all that empty land for a runoff.  Instead, it's going to head downhill . . . towards us.

My wife and I got on the phone to our insurers and discussed options.  FEMA's flood insurance isn't available everywhere, and is pretty basic - no cover for contents of your dwelling, outbuildings, etc. and restrictions on total coverage.  Flood insurance through private insurers can be much more comprehensive, but is also more expensive, depending how much coverage you want.  We ended up spending well over what FEMA would have cost us, but with much better coverage.  Premiums have to be paid in advance, which can be an unpleasant hit to one's wallet.  Also, FEMA makes you wait thirty days from the policy being issued before it takes effect;  private insurers usually only make you wait a week.  If you're looking for flood insurance at the last minute, that's something to bear in mind.

At any rate, I just thought I'd pass on this "heads up" warning to my readers.  I suspect some of us are going to need it over the next six months or so . . .

Peter

On the road again

 

My wife and Alma Boykin are both big Sabaton fans, and have been looking forward to their next live appearance in Dallas for months.  We're on the road today, headed for the Metroplex, where I'll be their chauffeur and dogsbody tonight.

No blog for today, but I'm sure you can amuse yourselves with the bloggers in the sidebar.

Peter

Confusion all round

 

I'm cynically amused at the reactions popping off like fireworks all over the political scene after the results of Tuesday's mid-term elections.  I expected the Democratic Party to do well, although not as well as it did.  I expected the Republican Party to lose ground, but not as much as it did.  The signs were not hard to see, and not hard to read.

Unfortunately, this means many topics of importance to conservatives, particularly the transsexuality brouhaha, the battle for Second Amendment rights, the reform (or otherwise) of our political and judicial system, etc. are going to be very much in play during the second half of President Trump's second term in office.  Sure, he has a veto, and he's threatened to use it:  but if he does so every time a bill reaches his desk, his political opponents will (quite rightly) portray him as a man who cares nothing for his country, but only for getting his own way.  Similarly, if his opponents try to bedevil him by passing bills that are in direct opposition to all he stands for, they need not be surprised if he refuses to play their game.  There's going to have to be compromise, and a lot of it, if our nation's government is to fulfil its purpose.  Can compromise be achieved?  One wonders . . .

I fear greatly that a culture of revenge may take root.  Block us when we're a minority?  Then we'll do exactly the same to you when we're a majority.  Work in an enterprise to which we're ideologically opposed (cough*SpaceX*cough)?  Then we'll do our best to undermine and overrule your enterprise at every opportunity.  Both sides have been guilty of this in the past.  If Democrats try to do it again, in a country as bitterly divided as is ours right now, sparks may fly - literally.

I dropped in at a local gun store on Wednesday morning, deliberately, to try to assess the reaction of staff and customers.  I was the second person in line, and within half an hour there were eight or nine more.  Almost without exception, they all wanted to look at - and many bought - the suppressors on display at one counter.  They all expect the Democratic Party to try to legislatively re-enact the tight controls over suppressors that were suspended a few months ago.  They were making sure to buy theirs, even if the models available weren't exactly what they wanted or cost more than they could immediately afford, for fear they might not be able to do so in future.

While we were there, the store owner received notification that two suppressor manufacturers had just implemented steep price increases;  and he told us that others had said they would not be able to increase production in a hurry to cater to a sudden increase in demand.  What's in stock right now will have to last, with some additional production, until the New Year.  I don't think that gun shop will have any suppressors at all in stock by Thursday morning.  Ammunition, too, is selling fast.  SGAmmo, my favorite online vendor, announced yesterday:

"We have seen a surge in ammo sales this past week, with a massive increase in the number of consumers placing orders, as well as the size of those orders, with daily sales numbers double to three times normal over the past 7 days and demand continuing to rise, reaching its highest point since March or 2026 ... With nothing expected but more factory price increases in the months ahead, and very significant potential for demand to spike, I can say with certainty that it would be a wise time to stock up. Conditions and pricing in the ammo market are not going to get better any time soon as high copper and energy prices continue to translate into increased manufacturing costs and reductions in stockpiled materials and products. If demand continues to stay high for much longer, there would be major shortages on popular ammo products very soon."

Their warnings have proved prescient in the past.  Fortunately, I don't need to worry.  I've already got mine.

Uncertainty always does things like that.  Stock up on what one considers essential, for fear the "others" may make it less available, or not available at all.  I guess we'll be seeing a lot of that for a while.

Peter

EDITED TO ADD:  From SGAmmo this morning:

"Yesterday, demand reached its highest point since March or 2020 when Covid-19 kicked off the biggest rush to buy ammo in history. It would be wise to buy sooner than later while we still have good options in stock. At this point I can almost guarantee that there will be a series of factory price increases in the weeks months ahead with many of the major factories following the lead of others. If demand continues to stay high for much longer, there would be major shortages that will last months to years before the market can be replenished."