Posted on 10/06/2026 9:09:39 AM PDT by kvanbrunt2
Google (GOOG, GOOGL) on Tuesday said it has inked a $4.3 billion deal with Constellation Energy (CEG) to bring 890 megawatts of nuclear energy online via a 20-year power purchasing agreement.
Constellation Energy stock climbed more than 12% on the news. Google parent Alphabet stock was flat.
(Excerpt) Read more at finance.yahoo.com ...
Dear FRiends,
We need your continuing support to keep FR funded. Your donations are our sole source of funding. No sugar daddies, no advertisers, no paid memberships, no commercial sales, no gimmicks, no tax subsidies. No spam, no pop-ups, no ad trackers.
If you enjoy using FR and agree it's a worthwhile endeavor, please consider making a contribution today:
Click here: to donate by Credit Card
Or here: to donate by PayPal
Or by mail to: Free Republic, LLC - PO Box 9771 - Fresno, CA 93794
Thank you very much and God bless you,
Jim
We can’t let Google acquire nuclear weapons.
Venezuela, Iran, eventually Russia — all point toward increased future availability of petroleum. And nuclear plants are coming on strong and will spread and provide green energy without petroleum. Energy is the lifeblood of modern economies and the future looks bright to me.
And my current view of AI is that we either get snuffed out in the next few years (I see this as a low probability) or else we are looking at a long future of prosperity and happiness (high probability in my opinion).
Actually, AI only knows from the data it has been trained on. You can train an AI with bad data and it will do what I have seen where if you ask how many sexes there are, it gives a wishy washy answer where, based on psychology, its a multi-dimensional (non-binary) answer.
AI trained with only facts will use those facts to reply to queries. But each AI creator can do things under the hood that the user has no insight into. I could see commies loving AI as a means to redefine what is fact and what is not.
You should not blindly trust artificial intelligence because it can generate false information, repeat biases, and sound completely confident even when it is wrong.
“AI. One is the engines require facts real data for the data centers.”
Wrong. AI is full on opinion and facts. And the facts and opinions are incomplete.
The DATA (which is both fact and opinion) can be broken down into what was retrieved from Harvard library but not Liberty library type of situations, what was retrieved from Brenan Center for Law and Justice but not Heritage, what was retrieved cheap, free or copywright easy to bridge but not what was omitted because it tightly protected by copywright.
Of course fact and opinion can be broken down into level of certainty of fact, concensus of fact, popular opinion, unpopular opinion, etc, etc.
Just an example. Some AI say GA has 161 counties (because that is how many the US Census says it has). And some say it has 159 counties (as that is how many GA government for 80 years says it has). And some AI miss some of the 159 counties because they saw a list that grouped 2 counties together with a hyphen.
If the person asking AI the question knows the arcane history of why the number of counties in GA changed he can word the AI question in a way that travels to the correct answer.
If the person is lucky he can get a B on the test, but not an A.
Ask an itelligent question, get an intelligent answer. But if you have to ask AI a simple question, how intelligent are you?
Jokes on them. We only have about four years left.
Oh, c’mon. Google is owned by Alphabet. Alphabet is owned by the CCP. If Google wants to nuke a US site they just have to pull one out of their Silicon Valley basement and deploy it.
I have a question. How exactly will AI “snuff” us out. I mean will it shoot us, poison us, starve us, blow us up? I don’t understand how AI is going to kill 8 Billion humans.
Most likely AI will get humans to do it for them.
For the past 50 years, we’ve said “no thanks” to a source of electrical power that would have made the Middle East oil ticks practically irrelevant.
This is a small step in the right direction.
There’s a degree of irony that the center of the Silicon Valley liberal hive is the driver behind the fission renaisance.
Did you not read the article?? It’s only for twenty years and then they have to give the nuclear weapons back. Try reading for comprehension.
And lets look at this as well :
“The Three Mile Island nuclear power plant may be restarted so its electricity can be sold to Microsoft to power data centers. “We want Microsoft to buy their electricity from zero-carbon energy sources instead of from coal plants, so it is in the interest of all of us that this nuclear power plant gets reopened,” states John Parsons (MIT Sloan).”
There are no major AI programs which include a "certainty rating scale" of the kind that you suggest. Maybe such a feature should be added to them.
Meanwhile, AI is very good at matching and extending language patterns. That does not mean the results will be correct. The results are very dependent on the quality of the training data models the AI uses. AI programs are designed to produce an answer. The program cannot determine if the answer is correct in real-world observations, only that it is logically consistent with the query and the training data model.
AI systems deteriorate very quickly if AI results are cycled back into the training model. That is because errors in the original data sources, plus natural language inconsistencies lead to incorrect results and cycling them back into the model magnifies the probability of subsequent errors.
AI is often wrong, but never in doubt.
This line of understanding of AI/SI is needed by all from Altman to the kindergartner.
Just because the King, or Scientist SAID it does not make it true or relevant. After WWII we emphasized IN GOD WE TRUST and UNDER GOD...not because we suddenly became more Godly...but because we realized that trusting in Stalin, or communism, or the Scientists of Germany, or any human or human artifact can be useful...but with major flaws.
Reagan tempered it with TRUST BUT VERIFY.
We can find SI/AI useful as a tool. But SI/AI is not to be trusted.
Yes, SI/AI data needs to be categorized, just as pre-AI data in computer systems is categorized.
There is reference data:
There is interpretation of reference data that is commonly accepted and backed up by other data in a pattern of consistency.
There is interpretation of reference data that is outside the mainstream, either because it is new, or because it was not supported by a consistency of patterns.
There is .....
LLMs are built on the underlying concept to not assume all its trained data are facts. This is why responses from AI models give you a probability of an answer. It’s good because it leaves the door open to new answers. The negative is its chance to hallucinate.
Hallucination? Also known as the MMIHO effect. Computer scientists discovered this early on. Magic Mushrooms In Hallucinations Out.
ill reply that for real ai to work forwarding knowledge it will need to be based on facts.
It is a FACT that Karl Marx had an opinion.
It is a FACT that Donald Trump had an opinion.
Opinions are needed in SI/AI. But they should be represented as opinion, and not as fact.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.