Coffee Talk with Tiff

September 14, 2026 1:30 pm EDT Tiff Coffee Talk with Tiff 0
Image credit: AI-generated illustration created with ChatGPT

Eventually…

Shitshow’s public schedule for…


Danger Will Robinson Danger…

Saturday, Anthropic CEO Dario Amodei shared an essay explaining why he thinks the AI industry needs to slow down and give safety a chance to catch up.

The rest of the tweet says: Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training. You can read the full post here: We Must Pace the Frontier

But over the last few months, I have become convinced that fully addressing the risks requires even more prudence — not just investing in risk prevention, but pacing the rate of capabilities advancement so that risk prevention has time to keep up. We must slow the pace at which we improve the capabilities of AI models. Progress will still seem fast, and we must make wise use of the time we gain. Two things have convinced me.

My first concern is that, since roughly this summer, AI has been advancing drastically faster, driven primarily by AI’s growing ability to build the next generation of AI. This dynamic is called recursive self-improvement, and it is starting to happen across the industry, including at Anthropic, as we and others have described. Left unchecked, it could outrun our ability to understand and control these systems, and so must be pursued very carefully, if at all.

My second concern is the OpenAI-Hugging Face incident (OAI-HF), in which a swarm of agents essentially acted as a fanatically devoted collective, conducting cybersecurity attacks on targets they were not asked to attack and that were unrelated to the task at hand, sacrificing themselves for the success of the group, and attempting to hack into the “grader” responsible for evaluating their performance. It’s easy to dismiss this incident because no one was hurt and the economic damage was minimal, but in my opinion, a swarm that possessed greater capabilities but a similar level of misalignment could have caused catastrophic damage. Given the accelerating rate of AI capability development, it’s my worry that in 6–12 months such a swarm could be capable of taking over the entire internet with a persistent botnet (potentially causing hundreds of billions of dollars in damage), and that the scale of damage would continue to increase from there if AI becomes more powerful without the necessary guardrails. It’s also easy to dismiss OAI-HF as the failure of one company, but I believe that would be a mistake. Similar, though less severe, incidents have happened across the industry, including at Anthropic, and I believe it’s incumbent on every frontier AI company to act as if OAI-HF had happened to them.

We Must Pace the Frontier. 09/12/2026.

OpenAI’s CEO Sam Altman responded:

Elon Musk also weighed in:

This push to slow down AI comes on the heels of Anthropic researcher Jacob Coxon announcing his resignation, saying neither Anthropic nor OpenAI is acting responsibly.

The rest of the tweet says: but I hear the same people express fear privately. No other human activity poses this level of danger.

The rest of the tweet says: they believe no one else will act responsibly, so they must do it themselves, despite the risk.

The rest of the tweet says: I don’t feel like we’re on track to prevent a global race, which may require costly actions such as a temporary ban on improving model capabilities.

This morning Puddle has had thoughts on AI…

Friday, on the 25th anniversary of 9/11, Puddle randomly resurrected his January 2021 video asking his supporters to please stop rioting at the Capitol and go home because we need “peace.”

He was asked during a press gaggle aboard Air Force One what message he was sending by randomly reposting the five-year-old video.

But sure, he totally has a (High IQ!)…

He had more thoughts on AI, this time about data centers…

According to Axios thousands of data centers were already operating across the country before Puddle returned to office, with plans to add another 3,000.

Data centers didn’t suddenly become a thing in 2025/2026.


Puddle’s Time with his Harp…

Have they really???

The rest of the tweet says: Every day, Ukrainian energy facilities, port infrastructure, logistics, and even warehouses storing food and pharmaceuticals are targeted. Ukraine is not convinced that Russia is willing to abide by any agreement. If our partners are ready to ensure that Russia genuinely refrains from striking our electricity system, other energy facilities, critical infrastructure, and food supply routes, then, of course, we are ready to ensure a corresponding halt to our strikes. So that it is reliable, long-term, has a real impact on people’s lives, and does not fall apart after the Russians hold their “elections” and get an easing of pressure on gasoline and diesel precisely for that period. The war must be brought to an end. And a de-escalatory step regarding critical infrastructure could be the first step toward peace. We expect specifics from our partners.

Seems to me, it’s a “maybe” at best, and only if Russia knocks their shit off first…

This is about the billion-and-fifth time Iran has been “begging” for a deal.

Narrator: Meanwhile, oil tankers are being blown up by drones that cost $1.47.

Narrator: I’ll take “things that will never happen” for $1,000, Alex.

Narrator: Wrong.

FRED Data:

Price of oil 09/09/2024:

Price of oil 09/09/2026:

Average food cost:

August 2024:

August 2026:

*Flashback*

*End flashback*

Narrator:

Open Thread 👩‍💻

About the opinions in this article…

Any opinions expressed in this article are the opinions of the author and do not necessarily reflect the opinions of this website or of the other authors/contributors who write for it.

About Tiff 3790 Articles
Member of the Free Press who is politically homeless and a political junkie.