Home | Search

2026-02-26

Update

• Very important. Read this: Risk disclaimer for having discussions with me on cyberattacking US govt
• I no longer endorse some of the claims below, and I haven't gotten around to updating them. I will write a cleaner less messy document, once I have some more clarity myself.
  • If you can help me get some clarity here, that would be nice actually.
• copy-pasting from somewhere else:

If you are working at or investing >$100M at an AI company building ASI, you are my enemy and you should not trust me. If you are some random unimportant and powerless person, you probably don't have to worry about me being violent or violating your consent or similar. I am seriously looking into plans that break US law, so there could be risks to you if you closely associate with me.

2026-02-16

Update

My thinking on "morality" and "status" seem to have undergone updates since I first wrote this post.

• I still am not happy with this post. I would like to have more clarity on exactly whose trust I wish to earn, and why, and what specific information would help to do that. Right now this seems like spam of lots of information, and in practice many of the people whose trust I wish to earn are too busy to sit and evaluate this much info.
• The far bigger problem with this post is I myself seem unhappy, and this might (or might not) be a result of some of the moral lines I endorse crossing
• It seems worth it for me to maybe invest more into having internal moral clarity first, before trying to also get anyone else to trust me completely.
• See also: Deliberate status engineering is hard

2026-02-12

Should you trust me?

Disclaimer

• Quick note, Incomplete
• Intended audience - anyone trying to figure out whether they should long-term collaborate with me or not.
• Contains politically sensitive info

Summary

• Watch my youtube channel for video interviews. Video contains implicit signals via body language.
• You can read my personal website.
• I have long track record of meaning what I say, and doing what I say I will do. I am open about believing in many unpopular ideas.
• I support "greater good justifies some harm". I support some actions that are illegal as per US law. I support some actions that some consider morally questionable. These are valid reasons for you to not want to associate with me too closely if you live in the US geopolitical sphere. Use your own judgment.
• MBTI and Big Five personality test results attached below.

Values and proof of trust

• Watch my youtube channel maybe, for example my interview while on hunger strike
  • Videos carry some implicit signals via body language, that are helpful for determining trustworthiness, which are not discernible via text.
• You can read my personal website.
  • I don't actually know if this is useful to you. It is not written for you, and it's definitely not written with the intention of earning your trust or similar. But I have personally observed that I seem to trust people more if I also know parts of them that they consider vulnerable to them, apart from work.
• Integrity
  • I generally mean what I say, and do what I say I will do.
  • Historical track record of high integrity, see for example internet forum posts since childhood, or my previous educational and work experience, or my financial track record.
  • I use disclaimers when I'm exploring ideas I'm not yet certain about.
• Openness
  • I have lots of opinions that society considers unpopular and am fairly open about it.
  • I am generally not willing to lie about this even if offered significant reward for doing so. (I think some people who are willing to hide their opinions for money are directly leading to the end of the world.)
  • This is a calculated decision on my part. I am aware there are pros and cons.
  • If I do acquire significant attention, I might choose to only signal boost some ideas and not others. But even so, I'm generally not actively hiding the other ideas, though I might not signal boost them as much.
• Moral values
  • No criminal record
  • Have significant social circle who can attest to my personality traits
  • I don't fully support liberalism but I'm more-or-less liberal in my day-to-day social interactions. Unless you wield a significant amount of power, most probably you are not a threat to me and I am not a threat to you. If you wield a significant amount of power, then it is more complicated.
  • I am openly willing to harm the lives of many people in order to pursue political objectives. I generally lean towards "greater good justifies some harm".
    • I have no moral issue with spreading fear and panic in general population due to AI risks, even though this comes with downsides such as poor mental health and suicides in general population. I don't know what rhetoric is most persuasive, but I think content creators should find out what rhetoric is most persuasive and use that. (I may have previously made twitter/youtube comments in this direction.)
    • I support public shaming of people working at labs building superintelligence. I don't think I am all that morally superior than they are, but I think the public thinks heavily in terms of status, and will only coordinate against at an outgroup it feels morally superior to.
    • I support leaking info of senior people building superintelligence, that makes them look bad in the eyes of the public. I support cyberattacks on them as well. (I am not working on cyberattacking anyone as of 2026-01.) (I have previously worked on a project to support whistleblowers.)
      • This is illegal as per US law, but I openly advocate for it anyway. This is actually a valid reason for you to not associate too closely with me, assuming you live in the US and seek power yourself.
      • I have also speculated in the above post, on cyberattacking various other actors (besides heads of AI companies and US govt) and leaking various other types of info (besides redacted info most relevant to AI risk). I don't yet have an answer I can long-term commit to, on which info about whom do I endorse leaking without consent.
    • I do not support threats of physical violence or actual violence against people working at AI companies buidling superintelligence. I am not planning to assassinate anyone, and I am not recommending people around me to assassinate anyone.
    • That being said, I don't have some principled stand against assassinating people. I think there are both pros and cons to assassinating people. ~I am currently not willing to discuss this topic in public.~
      • (Update 2026-02-12) I have now written posts speculating about assassinating AI company CEOs, atleast in theory. I still don't endorse actually trying to assassinate them in practice.
• Personality traits and values
  • See also: My personality test results
    • I have published my personality test results on various scales like Big Five, MBTI, and so on.

Collaboration

• If you agree with most of my beliefs and plans
  • We can hopefully collaborate a lot
• If you disagree with me on beliefs or moral values, but think I have integrity
  • Limited collaboration may be possible. I also accept anonymous email.
• If you aren't convinced I have integrity
  • Let me know what datapoints (about my past or future) could change your mind. I am aware the info above is not sufficient for me to win everyone's trust.

Subscribe

Enter email or phone number to subscribe. You will receive atmost one update per month

Comment

Enter comment