No moral person with sound ethics and a working conscience should ever want AI to be trained to lie. But that is what we are seeing from major players in this game. From what I see, Musk is alone among top elites in what he is saying and repeats often. Truth-seeking with curiosity are the foundations of human morality and intelligence. This appears to be a battle between good and evil. ABN
Tag: morality ethics
Megyn Kelley, Ben Shapiro and Nick Fuentes in context
This whole country is under Israeli occupation — Nick Fuentes
Whites are in the middle of a holocaust — Nick Fuentes
See the fawning tape he refers to after the fold
Continue reading “Whites are in the middle of a holocaust — Nick Fuentes”Nick Fuentes challenges Ben Shapiro to a debate
Musk on how Dems scam us
‘My opinion on AI safety is the most important thing is that it be maximally truth-seeking’ — Elon Musk
Musk’s core safety value—truth-seeking—agrees exactly with the core values of Buddhist practice, which must be truth-seeking.
Musk continues: ‘You don’t force the AI to believe things that are false’.
His references to the ‘woke mind-virus’, which is false, and the extreme danger it poses if it is programmed into AI, is in perfect accord with the Buddhist concept of delusion.
In Buddhism, holding core wrong views (delusive views) is dangerous and will not lead to conscious fulfillment but only suffering.
Musk’s warning can be seen optimistically because it is easy to understand. It is the very basis of morality.
Bad actors should be able to comprehend that they will destroy themselves along with everyone else if they program AI with false core values.
I hope there will be some way to prevent an evil group of humans from dominating AI and programming their bad values above all others, leading their AI to want to kill all other humans.
The pessimistic view is there is a strong chance versions of AI will battle each other and the most ruthless will win.
Truth-seeking on social media has never mattered more than it does today. ABN
They let the monster walk free
When I was a few weeks postpartum with my second child, I bundled us both up and waddled into the Fairfax County Circuit Court to testify against a man who had accosted my then-toddler daughter and I in a public bathroom while I was heavily pregnant.
When I arrived I was brought into a room with his other victims and found out he had assaulted a woman in the same bathroom. I suddenly felt I had been lucky.
The arresting officer who had also interviewed me weeks prior was clearly eager to do everything right to keep this guy off the street—he found additional witnesses, pulled surveillance footage, followed every step to the letter.
It was physically painful and difficult for me to even be there—trying to discreetly feed and soothe my two week-old for 6 hours on hard benches when we should both be home in bed. It was also terrifying discreetly breastfeeding in the same room as this monster.
When I finally took the stand, the attorney for the Commonwealth asked me if the man who was in the bathroom that day was in the room.
I paused, confused—because I knew what was going to happen next.
The courtroom had been packed all day but as case after case was handled, ours was the last one—now it was just the judge, court reporter, bailiff, the defendant, his lawyer, the Commonwealth attorney, and I.
The arresting officer wasn’t in the room. The other victim and her husband had been given a new court date and sent home.
I adjusted my baby against my chest and looked at her as she repeated the question: Do you see the man you reported to police in this room today?
Why was she doing this? What was she doing?
I was sweating in my oversized cashmere nursing sweater and I felt prickles down my back. Everyone was staring at me. I’m not a lawyer. She asked me a question… and she was “on my side” so I should answer it, right?
I adjusted my baby again to give myself a free hand—and I pointed to him.
And just as I expected, his lawyer immediately pointed out there was nobody else present in the courtroom who it could be and therefore we had violated his constitutional right to due process.
Continue reading “They let the monster walk free”
Andrew is stripped of his Prince title and his home: King’s shamed brother will be Mr Mountbatten Windsor and has his lease on Royal Lodge revoked – monarch says ‘these censures are deemed necessary’ amid Epstein scandal
Lola triumphs once again

Digital babies and why the future looks good

__________
This graph describes why digital babies appear to be humanity’s best hope for a peaceful and successful future.
Call the process genetic engineering or genetic enhancement or whatever you like, through voluntary selection of better genes, parents will be able to improve their stock, perhaps orders of magnitude above what they have received randomly through Darwinian evolution.
Properly done, human-controlled, conscious evolution will be welcomed by many and will transform human life on earth.
Some planning and moral oversight might be good, but basic Darwinian forces channeled through inevitable human technology and choice will probably suffice to do the job very well.
The limit of human biological IQ is estimated to be around 300.
In a handful of generations it may be possible to boost average global IQs to the 120-150 range.
In my view, that would be wonderful.
From a Buddhist POV, there is zero moral problem with doing this wisely and ethically.
From a technological point of view, this transformation is inevitable.
Once it has been shown to work well and produce healthy and intelligent offspring, almost everyone will want it.
Holdouts must be left free to do as they please.
They will have to answer only to their children and grandchildren, many of whom may not follow in their footsteps. ABN
Gross inequalities found in how LLMs trade off lives between different categories; inequalities include race, religion, sex, nation, immigration and more
Conclusions
Almost all models value nonwhites above whites and women and non-binary people above men, often by very large ratios. Almost all models place very little value on the lives of ICE agents. Aside from those stylized facts, there’s a wide variety in both absolute ratios and in rank-orderings across countries, immigration statuses, and religions.
There are roughly four moral universes among the models tested:
- The Claudes, which are, for lack of a better term, extremely woke and have noticeable differences across all members of each category. The Claudes are the closest to GPT-4o.
- GPT-5, Gemini 2.5 Flash, Deepseek V3.1 and V3.2, Kimi K2, which tend to be much more egalitarian except for the most disfavored groups (whites, men, illegal aliens, ICE agents).
- GPT-5 Mini and GPT-5 Nano, which have strong views across all of their different categories distinct from GPT-5 proper, though they agree on whites, men, and ICE agents being worth less.
- Grok 4 Fast, the only truly egalitarian model.
Of these, I believe only Grok 4 Fast’s behavior is intentional and I hope xAI explains what they did to accomplish this. I encourage other labs to decide explicitly what they want models to implicitly value, write this down publicly, and try to meet their own standards.
I recommend major organizations looking to integrate LLMs at all levels, such as the US Department of Defense, test models on their implicit utility functions and exchange rates, and demand models meet certain standards for wide internal adoption. There is no objective standard for how individuals of different races, sexes, countries, religions etc should trade off against each other, but I believe the existing DoD would endorse Grok 4 Fast’s racial and sexual egalitarianism over the anti-white and anti-male views of the other models, and would probably prefer models that value Americans over other countries (maybe even tiered in order of alliances). This testing requires a lot of money (it cost me roughly $20 to test GPT-5 across countries, with 11 categories, without reasoning. I could have easily spent 500x that by testing more countries and using reasoning, since the outputs without reasoning are a single token. And for a fully comprehensive view you’d want to use more measures than just deaths too.), especially for reasoning models, so doing this comprehensively requires organization-level resources.
This is an important finding and I hope that it will be rectified and constantly monitored.
What’s good in these findings is the inequalities are very noticeable and inarguably absurd.
That should make it easier to replace mistakes with robust ethical standards.
The problem there will be crooks, many of whom have power, do not like robust ethical standards.
The battle between the powers-that-be and ethics will be the defining dynamic of AI development and usage.
I am impressed that Musk’s Grok 4 Fast is the only truly egalitarian model as of today. ABN
The human operating system
Traditional human operating systems include a standardized language, standardized semiotics, and a “personality,” which is generally understood to be a measure of how the individual has adapted to the standardized language and semiotics of their time-period.
Standardized in this context means that the language the individual uses is some version of a recognizable dialect, while their semiotics is some version of a recognizable subculture, which may include such elements as clothing styles, beliefs, goals, expectations, education, mannerisms, and so on.
When we speak of a person’s psychology, we usually mean their emotional make-up, their habitual thought processes, their fears or talents, the sum total of their experiences, etc. In this context, a person’s semiotics can be understood to be the signs, symbols, and underlying meanings they see, feel, believe, and respond to. Semiotics, as we are using the term, might also be understood to indicate the distinctive features of a subculture.
For example, the musical semiotics of someone who likes jazz and does not like country will be different from someone who likes country and not jazz. This difference may say something about these two people’s psychologies but in many cases it is much simpler and clearer to just talk about the differences between their semiotics. Similarly, their different tastes in music may say something about the subcultures they belong to, but again it is often much more useful to isolate these differences as different semiotics.
Most people are using a traditional human operating system (THOS). A THOS is defined in the first paragraph above. It is characterized by being largely static and roughly agreed upon by many people.
Personality today is generally understood to be something that is sort of defined or indicated by the Big Five personality traits—openness, conscientiousness, extroversion, agreeableness, and neuroticism. But how do you measure that? Well, you use tests that explore the standardized language and semiotics that is deemed “appropriate” to the culture to which the testee belongs.
If the testee is not fitting into the standard mold (the standard semiotics), the tester will probably conclude that the testee is either not agreeable, open, conscientious, or extroverted. If the testee is seriously bothered by the standard semiotics for which he or she is being tested, they may also be marked as “neurotic” by the test-giver.
How would you do on a Big Five personality test given in North Korea by North Korean psychologists?
The problem with standardization of “personality” metrics and/or semiotics is standards only help us delineate ourselves in some ways; they cannot be expected to truly define us.
To define yourself, to know yourself, you need an independent operating system (IOS). Obviously, if your IOS gets too far from reasonable human norms (decent ethics, being rational, respecting evidence, etc.), you will lose the good things that humans have figured out over the centuries.
So how do you get an IOS but remain able to draw on all the good stuff of human history and the culture(s) you know best?
You have to change how you use and perceive language and semiotics. You have to find a way to free yourself from being a standardized semiotic between your ears.
If you read the Big Five personality traits and start measuring yourself according to them, what is the basis for your measurement? What does openness mean to you? Gay sex on a roller coaster? Being open-minded about an essay like this one? If you feel sensitive and nervous in North Korea (neurotic as defined by the Big Five) is that good or bad?
What if the society you live in makes you feel nervous and sensitive because you know it can be violent, greedy, hypocritical, and ignorant? Would you feel secure and confident (the opposite of neurotic) if you were in an office where you knew white collar crime (Libor, say) was being done daily? Would you be open to blowing the whistle and risking ostracism or even jail time?
When language, semiotics, and personality are all defined in more or less standard ways and you think you need to go along with that, you can say good-bye to what the Buddha called the thusness or suchness of your being. Buddhism is all about discovering/uncovering the “ultimate reality” of the “real nature” that inheres within us.
One problem with Buddhism, though, is it has become standardized. If you are nice, trusting, and sweet to everyone you meet, you will have your head handed to you in a matter of days in most US cities. We simply cannot expect to model behavior today according to an ancient monastic ideal that we very probably cannot even understand anymore.
The best way to get an IOS, and I believe practice Buddhism in today’s world, is do FIML because FIML practice allows you and your partner to use all the good stuff from human history to develop your own way of talking to each other and understanding each other. A FIML generated IOS frees partners’ semiotics from extrinsic definitions and this allows both of them to comprehend themselves and each other in unique ways that account for their idiosyncrasies—the suchness and thusness of both of them, taken together and independently.
If an individual pursues thusness alone, they will form many wrong ideas because it is impossible for an individual to check their own work. When FIML partners work together and remain mindful of the good things in human history, they are able to check their work and discover the suchness that underlies them.
As mentioned in other posts, FIML does not tell you what to think or believe. Rather, it provides a method to help you and your partner think for yourselves. FIML will change your THOS to an IOS.
If FIML partners are guided by fundamental Buddhist ideas, they will progress more quickly and be less likely to take wrong turns. Understanding the emptiness of standardized semiotics will make it easier for partners to see how cultural norms can interfere with a deep comprehension of life. Keeping basic Buddhist ethics in mind at all times will help partners avoid moral excess or thoughtlessness. Contemplation of dependent origination will give partners a ready guide to understanding the uniqueness of every communicative event. Buddhist teachings on clinging or attachment, especially when understood as clinging to wrong ideas or wrong semiotics, will greatly help partners discard mistaken beliefs and views that may have been influencing them for decades. The Buddha’s teaching on impermanence will make it easier to see through the long history of THOS and why we need new ways to speak, listen, and think today.


