A U.S. Senate investigation says the bank has continued hiding money stolen from victims of Nazism.

In the latest Senate findings, 70 Credit Suisse accounts with plausible links to Nazis in Argentina were opened after 1945 and at least 14 stayed open until the 2000s, including some as recently as 2020, according to the investigators’ reports.

The screen capture of a video (intersection approach starting at 9:00) shows clearly that Tesla’s bug-riddled software suddenly tries to veer left into the path of an oncoming truck (9:05), even though the route planner next to it has only a forward indicator.

If you blinked you might have missed that unexpected blue left arrow… and you’d perhaps then be dead. It’s unmistakable how this driver has less than a second to grab the wheel and force his Tesla back into its proper course and lane (9:06).

Why did version ELEVEN of software veer sharp left on a forward route with very obvious oncoming hazards?

Full video:

One theory seems to be after Elon Musk’s 2016 rushed demand that Tesla will depend on radar for safety, followed by Elon Musk’s 2021 rushed demand that Tesla will not use radar at all, he built a toxic culture of quality implosions and caused safety to decline with later models. This is the result of engineers being overruled by an erratic CEO’s “shiny-object” fantasy vision.

…system is randomly too conservative or too aggressive, i.e, can’t map out the surrounding correctly… I trust it as much as I trust R Kelly at a youth group.

The opposite of trust, the inverse of ethics. When you see any Tesla on the road, be prepared for it to assault others unpredictably. It’s a good reminder how society foolishly relies on Tesla drivers knowing, let alone wanting, to be risk adverse.

The latest regressive Twitter idea is a tax on public safety agencies of nearly $50K per month to continue providing emergency notifications.

On Friday and throughout the weekend, multiple National Weather Service (NWS) accounts announced that Twitter had removed their API access, which would disrupt crucial potentially life-saving automated emergency updates. The move came as Twitter prepares to transition its currently free API service to a paid subscription model starting at an exorbitant $42,000 per month for Enterprise access. […] “Since 2014, NWS has used Twitter’s API service to auto-post the latest warnings for tornadoes, severe thunderstorms, and flash floods to Twitter feeds that are followed by emergency managers, the media, and people in the path of dangerous weather,” the NWS said in its statement.
With Twitter’s new limits, the NWS will be unable to tell which automated emergency alerts go through and which don’t get posted. This will make the NWS Twitter accounts unreliable during weather emergencies. “For every warning issued, seconds could make the difference between life and death,” the statement reads, explaining how the automated emergency alert posts have an advantage over the forecasters who also tweet from the accounts.

People clearly will be killed if they keep depending on the Nazi-infested cesspool of Twitter.

It is shaping up to repeat the tragic levels of death we’ve seen among those who foolishly depended on Tesla.

To put it another way, in 2013 when a Tesla crossed a double-yellow and killed an innocent cyclist, Elon Musk announced abruptly an “autopilot” feature would prevent this crash. Ten years later Teslas crash far more than ever, still crossing double yellow and still killing cyclists. The announcement has all the hallmarks of an Advanced Fee Fraud system that never delivers the things promised to gullible buyers.

It’s worth noting that Elon Musk often leverages fears in the opposite way you might expect. He seems to relish in people worrying about tragedy because he allegedly sees a bunch of suckers he can defraud by forcing them to pay high prices, hooking them with premature announcements and false hope. I wonder if Twitter officers right now are excitedly licking their chops under a diabolical plan of rent seeking on emergency services especially during disasters.

Musk views information systems as coinage, a payment mechanism where he desperately tries to remove governance that inhibits rampant exploitation. He fraudulently calls government, public power as a form of representation, nothing more than a corporation for introducing errors.

That $50K per month tax for public service accounts is thus, for lack of a better word when describing pot-smoking libertarian product ideas… high.

Building a simple replacement for Twitter costs far, far less. Put two or three safety agencies together and you’re looking at Twitter trying to impose an grossly unjustified speech tax of over a million-dollars a year… just so basic text messages can post to a website?

Twitter also has become the kind of website that fewer and fewer people will want to visit, making it a less reliable site with far fewer people that somehow thinks it should inflate cost through the roof.

Interest in joining Twitter has plunged [81%] in the six months since Elon Musk’s takeover

The obvious answer is to leave Twitter with extreme urgency and spend a fraction of Musk’s stupidly high tax on speech for much higher grade public service communications infrastructure.

NWS for example could turn to a simple and elegant Mastodon example set by other public agencies, as it’s become a literal elephant in the room.

Here’s a big money quote from DYNOMIGHT in an assessment of cost to achieve AI scale for better intelligence (reduced loss):

…everyone reports that filtering the raw internet makes models better. They also report that including small but high-quality sources makes things better. But how much better? And why? As far as I can tell, there is no general “theory” for this. We might discover that counting tokens only takes you so far, and 10 years from now there is an enormous infrastructure for curating and cleaning data from hundreds of sources and people look back on our current fixation on the number of tokens with amusement.

10 years from now? There’s already a standard in infrastructure curating and cleaning data from hundreds of millions of sources: the W3C’s solidproject.org

Everyone says filtering data is better for intelligence? That’s nice to hear.

Nobody should believe in the absolute freedom of speech, when they factor high cost of predictable errors caused from data negligence.

How high? VERY high.

…the best current models have a total error of around 0.24 and cost around $2.5 million. To drop that to a total error of 0.12 would “only” cost around $230 million. …to scale a LLM to maximum performance would cost much more—with current technology, more than the GDP of the entire planet.

That 100X cost to reduce error seems prohibitive, although I’m sure someone is thinking $230 million is NBD like just one oligarch yacht.

DYNOMIGHT seems to be warning us current LLM architectures are about to run into a serious scale limit; they train already on all practically-available data with compute costs far too high.

Add integrity, gain intelligence at far lower cost to completely change the game. Serious food for thought considering Microsoft’s oligarchical ChatGPT has just been beaten by a pedestrian Web browser.

A month ago I asked Could you train a ChatGPT-beating model for $85,000 and run it in a browser?. $85,000 was a hypothetical training cost for LLaMA 7B plus Stanford Alpaca. “Run it in a browser” was based on the fact that Web Stable Diffusion runs a 1.9GB Stable Diffusion model in a browser, so maybe it’s not such a big leap to run a small Large Language Model there as well.

That second part has now happened.