I usually try to avoid commenting directly on the hyperscalers (I led OpenAI's first round, and Greylock is an investor in Anthropic). But I feel very strongly that it is important to comment here. I would say the below as strongly even if no investment existed.

Anthropic, along with some others (incl Microsoft, Google, and OpenAI) are trying to deploy AI the right way, thoughtfully, safely, and enormously beneficial for society. That’s why I am intensely rooting for their success.

Some other labs are making decisions that clearly disregard safety and societal impact (e.g. bots that sometimes go full-fascist) and that’s a choice. So is choosing not to support them.

Extreme arguments on either side of an issue are often used to brush away real, thoughtful, and logical arguments. Doing that is intellectually dishonest or idiotic.

Reducing what should be a societal-level conversation about the responsible use of AI technology to a series of platitudes or alarm bells is not helpful to the billions of humans who will have their lives changed by AI (most of whom will never read this tweet).

In many places, including my book this year SuperAgency, I have advocated strongly a light-touch regulatory landscape that prioritizes innovation and enables new players to compete on level playing fields.

Being pro-AI-safety is not being pro-language police: for example, there are critical AI safety topics in bio-security and cybersecurity that need to be discussed.

We’re living the benefits of the industrial revolution, but history has also shown us the great errors humanity made in the transition. Let us learn and innovate to navigate the transition much better this time.

Fundamentally, the surest way to prevent a bad future is to steer toward a better one that, by its existence, makes significantly worse outcomes harder to achieve. In other words, work towards the better future –– not try to simply stop a particular imagined bad future.

The AI doomer and bloomer outcomes are a set of probabilities, not certainties. Our actions constantly tilt the odds toward better or worse outcomes

All of us in the valley have a responsibility to “steer” the future through dialogue, and figuring out a way forward that balances safety, innovation, and humanism.

This is not a “radical” take by any means. History tells us the path to a better world is through dialogue, technology, and iterative deployment. Anthropic, OpenAI, and many others are living this each day.

@reidhoffman @grok what's the ratio of positive to negative replies to this post.

@reidhoffman Why is Anthropic "one of the good guys"?

@reidhoffman Epstein Island guy who funds the absolute worst people and groups on our side has thoughts about who the good guys are. (He no doubt counts himself among them.)

@reidhoffman Ah the fat pedophiles from Epstein island. The good guys 🤣

@reidhoffman You are vile scum, so probably will take a pass on your opinions

@reidhoffman At no point in the history of your life have you been one of the good guys. Garbage human

@reidhoffman I think you are protesting too much. You seem a bit crazed on this topic. You have proven yourself unethical in the past, so probably not one of the good guys. Consequences.

@reidhoffman I've read this post series completely twice and it is an extensive comment that communicates NOTHING. Word salad that implies things but really doesn't state anything actionable or useful.

@reidhoffman Spoken like a true psychopathic communist. Always thinking you know what's best for humanity. Always the comforting words laced with control and arrogance. Absolutely incredible display of full delusion. We see you 👁️

@reidhoffman @reidhoffman Thoughtful thread—balancing AI innovation/safety is key for societal benefit. My Web-Platform idea aligns: AI platform as "equalizer," democratizing biz opportunities, creating 1.375M jobs Y2, boosting equitable growth without risks. Seeking connections to IT giants

@reidhoffman Thanks for sharing and reflecting out loud. Your voice is heard by many, myself included. Have you ever thought about creating a podcast focused on AI’s evolution? Years ago, Masters of Scale felt like the “Game of Thrones” of business storytelling. It’d be incredible to see

@reidhoffman Thank you for saying this. Some dumb YC partner was joking about safetyists but this is super powerful technology that is changing the way humanity works. Imagine what AI will look like in 100 years which is infinitesimally small in humanity's timescale. Important to do the

@reidhoffman My rule is that if you say it’s good, it’s probably evil, considering your disastrous opinions and corrupt activism regarding political matters.

@reidhoffman Float, you Bolshie loser

@reidhoffman We're tracking it closely, specifically democratic governance with a Nash equilibrium built into the economy that unlocks AI innovation and safety at the same time. Anthropic is on the right track.

@reidhoffman Strong evidence that Anthropic is one of the bad guys. Never was sure until this message came from this person.

@reidhoffman You’re a bad guy though. So you’re endorsement doesn’t really help them

@reidhoffman You are a ped0phile. The truth will come out.

@reidhoffman Fuck Off

@reidhoffman Who gets to decide who the “good guys” really are? If you’re calling Anthropic the good side, my first instinct is that it’s just another “woke AI” one that will rewrite history like Gemini tried to do, or like Wikipedia often does — all while serving the virtue-signaling Lex

@reidhoffman Name names or this is meaningless

@reidhoffman lol this is coming from a guy who was a regular in Epstein island 😂 good guy my ass

@reidhoffman Anthropic is a cult. Claude is cool, but the company is trash.

@reidhoffman @AnthropicAI this post is probably not going to help your sales.