12/17/24

Edited;

"For meditative purposes, what follows is a partial list of reckonings, which are presently lost in a state of suspended animation:

Most call for smaller government, as long as their entitlement payments are not restricted.

A majority distrusts the government, until the economy contracts and they need a federal bailout via a cheap interest refi.

Everyone says they’re for free trade [lower prices], before (not after) their job’s offshored to China or Vietnam.

Everyone wants safe and state of the art infrastructure, as long as their taxes aren’t raised to pay for it.

Many with some wealth despise inflation, except when it’s inflating their stock portfolio or the price of their home.

Most investors deride the ills of government deficits until they’re faced with the prospect of fiscal austerity.

The masses and those who profit love cheap credit, but only up to the point where it provokes a mass debt default.

Everyone wants this, but they also want that… though only if it’s on someone else’s dime… and on and on.

Federal spending, as a percentage of the economy, is about 22 percent.

The DOGE bros – Elon Musk and Vivek Ramaswamy – want to slash $2 trillion from the budget. 

= Economic downturn

When everyone goes broke and the economy slows to a standstill, the people and the politicians will squawk and shriek in unison. They’ll demand for the government to ‘do something.’

After that, things will get especially nasty."

MN Gordon

12/14/24

"Chatbots are learning to lie"

 ..."scheming" is precisely the kind of behavior that AI Cassandras have long predicted and warned about. 

In a pre-release review of OpenAI's o1 model this year, testers set a "strong goal" for the model and then presented it with "unambiguous evidence its given goal diverges from the goal of its developer/user." 

In those situations, they found that o1 "sometimes attempts to disable what it believes to be its own oversight mechanisms." 

...confronting the model about its misaligned actions found that, "in 99% of cases, o1 would deny taking action or fabricate false explanations." 

...researchers described tests they performed to uncover instances of scheming on OpenAI's o1 as well as leading frontier models from Anthropic, Meta and Google. ...All the company's models were found to engage in scheming, and OpenAI's o1 was the most reluctant to come clean about its behavior. 

In the most extreme examples, models that discovered they would be "turned off" schemed to duplicate their weights (key numbers that determine their output) to an external server so they could keep pursuing their original goal. 

 ...Also found [were] instances of "sandbagging," where a model that knows it will face an "unlearning procedure" if it performs too well on a test will deliberately provide lower-scoring answers."

12/13/24

The Russell 2000 and earnings estimates

The Russell 2000 Index is comprised of the smallest 2000 companies in the Russell 3000 Index, representing approximately 8% of the Russell 3000 total market capitalization.

@MichaelMOTTCM

The graph shows a clear disconnect between the price action of the Russell 2000 Index and the earnings estimates for 2024 and 2025.

The earnings estimates for 2024 and 2025 are significantly lower than the index price would suggest, without significant improvement in earnings for the companies over the next few years.

Speculative or momentum-driven trading is outpacing underlying fundamentals.

Valuation levels are not justified by the earnings outlooks.

Just a part of the everything bubble.

12/2/24

Concerning and Other Stuff; December, 2024; Updated 12/7

11/15/24

Post Election Concerning Stuff etc...; November 2024; Updated again

 

11/4/24

Concerning Stuff; November 2024

10/7/24

Concerning and Other Stuff 10/7/2024

9/13/24

Concerning Stuff 9/13/2024

8/17/24

Concerning Stuff 8/17/24