Steven Marlow

Cylon enthusiast. Introspective Behaviorism. Innateness. Theorist of Mind. Ixian. Field of AI died in 2012. Loves robots, not lover robots. (cog/acc)

[Steven Marlow @sd_marlow]

AGI by Nov 18, 2027
*more of a self-challenge than anything else.
**won't use ML/DL tech tree

[Steven Marlow @sd_marlow]

@BrunoAcciolyAI I told you I'm not arguing for any of that, but you doubled-down. Regardless, neural networks don't have ANY functional elements that can be associated with self-awareness or cognition. It's not a memory. It doesn't store knowledge. Search is passive access.

[Steven Marlow @sd_marlow]

Can we please get this guy into the AI retirement home ASAP.
People that want to believe are going to eat this crap up, but it's a huge embarrassment for the industry, and makes him look like a n00b with psychosis.

[Alex Kantrowitz @Kantrowitz]

AI Pioneer Geoff Hinton tells me he believes AI is conscious.... and humans better get used to the idea that they're not the only intelligent life on earth.

"They've very like us," he says. "They're beings like us."

AI chatbots, he says, must understand your questions in order to answer them. There's an awareness there that equates to sentience. "We're going to have to accept that intelligence is not just biological."

[Steven Marlow @sd_marlow]

@BrunoAcciolyAI CoT is NOT a reflection of internal deliberation (for LLMs). It's just a new chain of outputs. Rather than have a mechanism for self-reflection, they have proven to be exceptional at reflecting the whims and desires of the people using them. A mirror, not a glass window.

[Steven Marlow @sd_marlow]

@BrunoAcciolyAI When they "read" text, there is no translation into "ground truth" where processing happens (based on knowledge and perspective, etc).
They are intercepting text, and statistically, drafting a text reply, with zero understanding of the meaning behind the text.

[Steven Marlow @sd_marlow]

@BrunoAcciolyAI Consciousness (self-awareness) is a specific aspect of mental activities, and "functional" is adding computational functions that are pre- and post- process, including aspects that we can't directly address/observe.
Reductionism strips context.

[Steven Marlow @sd_marlow]

Introspection-as-a-Service

[Steven Marlow @sd_marlow]

@atmoio Agree, but don't forget that scripting for these things is done at the written layer as well.

[Steven Marlow @sd_marlow]

@Scobleizer @BlissyOnX @Shmall Seems like you really get a device where every user has the potential to mod it, and have those mods distribute (for free). Crowd sourcing more than open sourcing.
*but people still need to know whats new; what takes-off. Where the ads will sit. ;p

[Steven Marlow @sd_marlow]

@Scobleizer "To understand the true value of poke, you need to have gone through the pain of bad human assistants & accountants" - was that actually in the pitch deck?

[Steven Marlow @sd_marlow]

@edzitron Brute-force/neural networks was NEVER the right path to be on.

[Steven Marlow @sd_marlow]

@edzitron Kosher technology.

[Steven Marlow @sd_marlow]

@BernardJBaars That limited scratch pad is why consciousness is kinda of a bad thing, because at that level, you're not working with all of the information.
Let your mind do it's thing.

[Steven Marlow @sd_marlow]

@BernardJBaars That limited scratch pad is why consciousness is kinda of a bad thing, because at that level, you're not working with all of the information.
Let your mind do it's thing.

[Steven Marlow @sd_marlow]

@chris_j_paxton Now we are addressing priors, common sense, and "general" intelligence in the correct way.

[Steven Marlow @sd_marlow]

@edzitron Like selling a video game before you've finished building it, but it's basically the entire market paying for development with no money left for actual sales, assuming you even finish the game (or if it's even any good).

[Steven Marlow @sd_marlow]

If someone is going around at night, looking for an unlocked car, you don't take money from the car owners to buy that person their own car!
Is AI making everyone THAT stupid?

[Rohan Paul @rohanpaul_ai]

FT: Trump administration, OpenAI discussing possible government stake in the AI startup.

OpenAI has already floated a Public Wealth Fund, which would hold long-term assets and potentially pass AI-related gains back to citizens. It will be a government-backed vehicle outside the company.

OpenAI said the fund could “invest in diversified, long-term assets” and would enable citizens to participate in the “upside” of AI growth, possibly by receiving the fund’s returns directly

Theres a huge political angle: voters fear job loss, data-center costs, and corporate control, while AI companies need Washington’s support for infrastructure, procurement, and regulation.

For some context, The Trump administration has already taken stakes in Intel, IBM and other quantum and critical mineral companies during the president’s second term.

---

ft .com/content/b1ab6106-77e6-4218-9eb4-e44bd56ca400?syn-25a6b1a6=1

[Steven Marlow @sd_marlow]

@atmoio Their valuation is a check reality won't be able to cash.

[Steven Marlow @sd_marlow]

@realOliverStark Representatives game the system when their party has the votes to do so. Legislative processes need to be more resilient in the face of power-seeking behavior.
reform reform reform

[Steven Marlow @sd_marlow]

@BrunoAcciolyAI If it helps you:
I feel the framework from which I draw my conclusion is "more right" than the framework Hinton is using to draw his own.
"Academic rigor," in this case, is the failure point.

[Steven Marlow @sd_marlow]

@BrunoAcciolyAI If you have the wrong definition of consciousness, then you just don't understand it. That so many, including people you've cited, freely mix terms like conscious, phenomenal consciousness, and being sentient, it's impossible to have an intelligent discussion on the topic.