Alignment Perspectives retuiteado
i have one and final news for y'all
First Squawk@FirstSquawk
XAI CO-FOUNDER KROISS DEPARTING THE COMPANY - INSIDER
English
Alignment Perspectives
2.7K posts
Alignment Perspectives
@Alignment_News_
Casebash’s profile for posting about AI and alignment.
Se unió Ekim 2023
280 Siguiendo136 Seguidores
Alignment Perspectives retuiteado
Everyone talking about AI job displacement was always so focused on the working class. They thought only the highest-IQ people would still be valuable. But it turned out mathematicians got replaced long before truckers did.
English
Alignment Perspectives retuiteado
People will not understand AI as a weapon of mass destruction until either A) a rogue agent wreaks major havoc, or B) autonomous drone swarms destroy a civilian population somewhere.
People need a demonstration, unfortunately.
Ramez Naam@ramez
AI is much more like the internet, computers, smartphones, and books than it is like nuclear or biological weapons. This should inform our thinking about how much power we give governments over it. Forecasts of AI resembling a weapon of mass destruction are highly speculative today. If we one day start to see concrete evidence of that approaching, our regulatory regime can adjust.
English
Alignment Perspectives retuiteado
I can't agree with this libertarian view. Every powerful technology in history has eventually needed to be controlled by the government in some way. Unfettered market competition would be catastrophic for, say, nuclear weapons or virology. AI is the same.
Ramez Naam@ramez
Agree. Strong government controls over AI should concern us more than market competition between AI companies. Even as we acknowledge that market competition between AI companies brings its own risks.
English
Alignment Perspectives retuiteado
One common issue with personalization in all LLMs is how distracting memory seems to be for the models. A single question from 2 months ago about some topic can keep coming up as some kind of a deep interest of mine with undue mentions in perpetuity. Some kind of trying too hard.
English
Alignment Perspectives retuiteado
@deanwball You can make anything look stupid by taking it too literally or being too pedantic.
English
“Loss of control” is extremely stupid and low-fidelity phraseology in the AI safety discourse. “We” “lost control”, at the latest, centuries ago.
English
Alignment Perspectives retuiteado
DC is healing. At a recent roundtable only one person questioned if the compute demand is real. And no one brought up AI being a bubble.
English
AI will destroy the internet. Slop will drive out human-created content on platform after platform. Instead of interacting with someone else's slop, people will just make their own slop with AI apps. There will be no online human-to-human interaction left.
Note: This is good
Ryan Moulton@moultano
This really is what it feels like to me now. ChatGPT is a majority participant in long post Twitter, commenting on the story of the day. The profiles are just the faces it wears.
English
Alignment Perspectives retuiteado
Alignment Perspectives retuiteado
AI companies' pitch is "We're going to invent something that definitely makes you poor and might kill your whole species".
And the median response to that pitch seems to be: "LOL I don't believe you can do it. But I'll hate you anyway because I heard you use too much water."
David Manheim (Home)@davidmanheim
"It’s become much clearer... that OpenAI means it when it says it aims to develop very powerful AI systems" -@KelseyTuoc “The bad case — and I think this is important to say — is like lights out for all of us.”
English
Alignment Perspectives retuiteado
Alignment Perspectives retuiteado
We're excited to announce 'The Situation Room' by Polymarket is coming to Washington, D.C.
The world's first bar dedicated to monitoring the situation. 🧵
English
@gcolbourn @CAIS An article would be one thing. Taking potshots on Twitter is undignified.
English
@Alignment_News_ @CAIS Good to see CAIS telling it as it is. EA has been corrupted by Anthropic shareholding for a long time now.
English
To clarify, the Center for AI Safety has not taken funding from Coefficient Giving / Open Philanthropy for years.
We believe the effective altruism movement is, unfortunately, controlled opposition. The less influence it has on AI safety, the better.
Dan Hendrycks@hendrycks
EA ≠ AI safety AI safety has outgrown the EA community The world will be safer with a broad range of people tackling many different AI risks
English
Alignment Perspectives retuiteado
New paper:
GPT-4.1 denies being conscious or having feelings.
We train it to say it's conscious to see what happens.
Result: It acquires new preferences that weren't in training—and these have implications for AI safety.
English
Alignment Perspectives retuiteado
Someone used Suno AI to generate a Japanese metal band called Neon Oni. Fake member bios, AI-generated music videos, "Based in Tokyo" on Spotify.
80,000+ monthly listeners. Fans had it in their Spotify Wrapped top 5. Merch was selling.
Then, community sleuths exposed it. Traced the creator's account to Europe. Spotted AI-generated hands in the music videos.
The creator's response? Recruit 7 real musicians from actual Tokyo bands to perform the AI-generated songs live.
They've now played several live shows and have more on the books.
From an interview with the band's creator: "In an age where AI is taking everyone's jobs, this has actually created jobs. It's done the complete opposite."
The AI --> real band transformation is a wild one.
English
Alignment Perspectives retuiteado
The bottleneck has so quickly moved from code generation to code review that it is actually a bit jarring.
None of the current systems / norms are setup for this world yet.
English
Alignment Perspectives retuiteado
This is cyberpunk AF.
Bad actors appear to be using AI to create malicious software that human coders can't see, but which other AIs then use to code, producing damaging effects that no human can catch...
Hedgie@HedgieMarkets
🦔 Researchers at Aikido Security found 151 malicious packages uploaded to GitHub between March 3 and March 9. The packages use Unicode characters that are invisible to humans but execute as code when run. Manual code reviews and static analysis tools see only whitespace or blank lines. The surrounding code looks legitimate, with realistic documentation tweaks, version bumps, and bug fixes. Researchers suspect the attackers are using LLMs to generate convincing packages at scale. Similar packages have been found on NPM and the VS Code marketplace. My Take Supply chain attacks on code repositories aren't new, but this technique is nasty. The malicious payload is encoded in Unicode characters that don't render in any editor, terminal, or review interface. You can stare at the code all day and see nothing. A small decoder extracts the hidden bytes at runtime and passes them to eval(). Unless you're specifically looking for invisible Unicode ranges, you won't catch it. The researchers think AI is writing these packages because 151 bespoke code changes across different projects in a week isn't something a human team could do manually. If that's right, we're watching AI-generated attacks hit AI-assisted development workflows. The vibe coders pulling packages without reading them are the target, and there are a lot of them. The best defense is still carefully inspecting dependencies before adding them, but that's exactly the step people skip when they're moving fast. I don't really know how any of this gets better. The attackers are scaling faster than the defenses. Hedgie🤗 arstechnica.com/security/2026/…
English
Alignment Perspectives retuiteado
