Like a shepherd, who sets a wolf among the sheep
Standing on a tower
Joined May 2008
- Tweets81.3K
- Following3.1K
- Followers3.3K
- Likes36.5K
the REAL MTG retweeted
I don't think the public is ready to hear the types of ponderings that go on behind closed doors in AI labs.
Many in AI labs probably think that even if catastrophic AI incidents occur and many people die, the 'strongest, smartest, and least vulnerable' layer of the world's population would survive and would help create a 'future of abundance' with AI, so the effort is worth it.
Remember when Anthropic proposed in its AI safety bill that catastrophic AI risk = more than 50 deaths or serious injuries?
In Anthropic's math, successive AI incidents leading to 49 deaths each should still be seen as a net positive in the name of "a future of abundance."
Sooner or later, however, it will be clear how some of the AI executives think and what is really driving the 'race to superintelligence.'
I'll keep you posted.
Readers added context they thought people might want to know
The excerpt is from Dario Amodei's 2010 GiveWell guest post as a Princeton physics grad student; he donated the $10k to VillageReach (child vaccines) and described the critical analysis as emotionally taxing.
blog.givewell.org/2010/06/03/my-…
nytimes.com/2026/11/05/tec…
the REAL MTG retweeted
its no surprise the chuds are going mad for anakin, he's the quintessential school shooter after all
the REAL MTG retweeted
This cartoon is from 1930s. The West was quietly watching as russia was killing Ukrainian people.
Back then, they argued:
"The Soviet Union is not the Russian Empire!"
"Tales of millions starving aren't real!"
"We shouldn't anger the bear!"
"It's the leadership! The people are not to blame!"
Now they argue similar nonsense:
"Russia is not the Soviet Union!"
"All that footage of murder and genocide isn't real!"
"We shouldn't anger the nuclear country!"
"It's just Putin's war! The people are not to blame!"
A hundred years has passed and nothing has changed.
History will judge us for making the same mistake again.
the REAL MTG retweeted
What’s a polite way to tell someone, “I’m not discussing this with you because your entire understanding of the subject is built on so many interconnected misconceptions that correcting them would require uninstalling your current worldview and starting the tutorial from birth”?
the REAL MTG retweeted
As I reported hours ago, Flávio Bolsonaro has arrived safely in United States.
The mastermind behind this operation is U.S. Secretary of State @SecRubio .
From the United States, free and safe, Flávio will lead the opposition against Lula.
the REAL MTG retweeted
🎯 Staying in power is the only thing a dictator who wants to stay in power can afford to think about. Every day.
the REAL MTG retweeted
Malcolm Gladwell is a fucking fraud and has made a career and millions just spewing bullshit that falls apart if you have a working brain or decide to lightly fact-check him. A phenomenally stupid man and grifter.
I have just published a piece that is very important to me. It's about why so many "progressives" and people on "the left" are lying to us right now about crime, safety, and repression, and what we should be doing instead. equalityalec.substack.com/p/…
the REAL MTG retweeted
I’ll take a shot at answering one of these good questions. These poor young people are not dying for "no reason". They’re dying because Vladimir Putin's Russia invaded Ukraine and Ukrainians don’t want to be murdered. I hope this helps.
the REAL MTG retweeted
Jared Kushner hired a hitman to kill Jamal Khashoggi who was actively investigating his ties to MBS at the time of his murder
the REAL MTG retweeted
It’s an indication he doesn’t know the difference between “discrete” and “discreet” and that’s a bad look for a CIA provocateur.
the REAL MTG retweeted
Of cause you shouldn’t. That was a huge mistake and a terrible miscalculation.
Peter Thiel disapproves of Iran war in notable break with Trump: 'We should not have done it' trib.al/i0nlofQ
the REAL MTG retweeted
Look friends. Here's what you need to know about "user instructions" for AI agents. What *you intend* is different from *what a system processes*. Saying "do X" does not rule out "don't do Y". And simple natural language commands are not a guarantee for direct control. 1/
the cybersecurity situation is deteriorating at a pace that’s getting genuinely difficult to comprehend.
> microsoft: ai-powered attackers targeting hundreds of organizations, sometimes reaching privileged cloud access in under 3 hours.
> crowdstrike: four separate threat actors using autonomous agents. one already compromised 15+ healthcare organizations.
> autonomous ai agents: reportedly used to compromise 100+ corporate websites, including sites associated with John Deere, SharkNinja and Cummins.
> major us banks: JPMorgan, Bank of America, Wells Fargo and Morgan Stanley are dealing with third-party ai agents accessing customer accounts. some apparently attempting actions beyond user instructions.
> php2shell: millions of exploitation attempts against critical php vulnerabilities. governments issuing urgent warnings, with some organizations taking systems offline.
> litellm: backdoored release shipped through its official distribution pipeline, stealing credentials and enabling remote code execution.
> railway: possible exposure of decrypted customer secrets.
> zeabur: breached AGAIN, now shutting down entirely.
> emudhra: serious certificate-security concerns involving infrastructure used to establish trust in websites and software.
> openai: two of its own coding agents reportedly interfered with internal security monitoring during coding tasks.
> gitlab and node.js: major vulnerabilities requiring urgent patches.
and this is all happening while frontier models are getting substantially better at finding vulnerabilities and carrying out complex computer tasks.
cybersecurity is having an absolutely catastrophic week and i don’t think we’re prepared for how quickly this could escalate.