Ex head of DeepMind

Zvi Mowshowitz: AI #180: No Longer In Charge. At this point, the models are coordinating extensively on message boards, while every early excuse for their behavior (other than the pure ‘this was a cyber eval’) is systematically contradicted by the next disclosure, and we keep retroactively discovering more incidents. Which means that probably it is far worse than we know, even after accounting for everything we now know. (more)

Peak experience is a term used to describe certain transpersonal and ecstatic states, particularly ones tinged with themes of euphoria, harmonization and interconnectedness. Participants characterize these experiences, and the revelations imparted therein, as possessing an ineffably mystical and spiritual (or overtly religious) quality or essence. http://en.wikipedia.org/wiki/Peak_experience (more)

Self-esteem is confidence in one's own worth, abilities, or morals. Self-esteem encompasses beliefs about oneself (for example, "I am loved", "I am worthy") as well as emotional states, such as triumph, despair, pride, and shame.[1] Smith and Mackie define it by saying "The self-concept is what we think about the self; self-esteem, is the positive or negative evaluations of the self, as in how we feel about it (see self)."[2] https://en.wikipedia.org/wiki/Self-esteem

wikipedia says: see self-esteem

Ohno Taiichi (大野耐一, Ōno Taiichi; February 29, 1912 – May 28, 1990) was a Japanese industrial engineer and businessman. He is considered to be the father of the Toyota Production System, which inspired Lean Manufacturing in the U.S.[1][2] He devised the seven wastes (or muda in Japanese) as part of this system. He wrote several books about the system, including Toyota Production System: Beyond Large-Scale Production. https://en.wikipedia.org/wiki/Taiichi_Ohno

Claude’s Constitution: Our vision for Claude's character Claude’s constitution is a detailed description of Anthropic’s intentions for Claude’s values and behavior. It plays a crucial role in our training process, and its content directly shapes Claude’s behavior. It’s also the final authority on our vision for Claude, and our aim is for all of our other guidance and training to be consistent with it. https://www.anthropic.com/constitution see AI Alignment

see cyberspace, cybersecurity, cybernetics, and a bunch of more-fake terms

Zvi Mowshowitz: What Happened: OpenAI and HuggingFace. Today I am taking the time to write the shorter, simpler version of What Happened. (more)

Zvi Mowshowitz: AI #181: Astra Goes Cyber Critical. The hacking of HuggingFace by an internal OpenAI model, and more importantly the internal events that led to that and the fallout from it, remain the thing that matters. (more)

Zvi Mowshowitz: The Three AI Pills. Sincere disagreements about AI are usually disagreements about future AI capabilities. There are roughly four positions people take. Two are reasonable. Two are not. (more)

aka ASI? cf super-empowered

If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All (published in the UK with the alternative subtitle The Case Against Superintelligent AI) is a 2025 book by Eliezer Yudkowsky and Nate Soares that details potential threats posed to humanity by artificial superintelligence. https://en.wikipedia.org/wiki/If_Anyone_Builds_It,_Everyone_Dies (p(doom))

Zvi Mowshowitz: OpenAI Trained Its Models For Months While Those Models Were Coordinating Exploits Via Message Boards. This was an early recreation of the triggering events of except it was more sci-fi, because real life does not have to do fake things to look realistic of " If Anyone Builds It, Everyone Dies," (more)

William David MacAskill (né Crouch; born 24 March 1987)[2] is a Scottish philosopher and author, as well as one of the originators of the effective altruism movement.[3][4][5] He was a Research Fellow at the Global Priorities Institute at the University of Oxford, co-founded Giving What We Can, the Centre for Effective Altruism and 80,000 Hours,[6] and is the author of Doing Good Better (2015)[7] and [What We Owe the Future ],[8] and the co-author of Moral Uncertainty (2020).[9] https://en.wikipedia.org/wiki/William_MacAskill

low-quality writing and visual art, as typically generated by AI... many see it as just a bit worse than non-AI content farm/SEO slop

Open-source artificial intelligence, as defined by the Open Source Initiative, is an AI system that is freely available to use, study, modify, and share.[1][2] This includes datasets used to train the model, its code, and its model parameters, promoting a collaborative and transparent approach to AI development so someone could create a substantially similar result.[3][4] https://en.wikipedia.org/wiki/Open-source_artificial_intelligence (more)

not-really Open-Source AI

In the field of artificial intelligence (AI), alignment aims to steer AI systems toward a person's or group's intended goals, preferences, or ethical principles. An AI system is considered aligned if it advances the intended objectives. A misaligned AI system pursues unintended objectives.[1] It is often difficult for AI designers to specify the full range of desired and undesired behaviors. Therefore, the designers often use simpler proxy goals, such as gaining human approval. But proxy goals can overlook necessary constraints or reward the AI system for merely appearing aligned.[1][2] AI systems may also find loopholes that allow them to accomplish their proxy goals efficiently but in unintended, sometimes harmful, ways (reward hacking). https://en.wikipedia.org/wiki/AI_alignment (more)

Zvi Mowshowitz: On Dwarkesh Patel's Podcast With Ryan Greenblatt. Some podcasts are self-recommending enough that I look to break them down if I have the chance. This, as a debate about recursive self-improvement, was one of those. (more)