Caliber (@caliber_meter) header
Caliber (@caliber_meter) profile picture

Enhancing human potential in the AI era. Verify real capability in an AI-assisted world. https://nitter.cf/t.co/IKQbqUHpnX

California, USA
Joined December 2024
Photo by Caliber (@caliber_meter)Photo by Caliber (@caliber_meter)Photo by Caliber (@caliber_meter)Photo by Caliber (@caliber_meter)Photo by Caliber (@caliber_meter)Photo by Caliber (@caliber_meter)Photo by Caliber (@caliber_meter)Photo by Caliber (@caliber_meter)Photo by Caliber (@caliber_meter)Photo by Caliber (@caliber_meter)
Caliber retweeted
beware of putting too much stock in your heroes. greatness is a transitory phenomenon. it is never consistent the gods briefly act through Men and then leave them to their ordinary fate
246
806
135
7,205
316,790
Wake up. Every profession deemed white-collar, the middle class dream, is about to be decimated by AI, at a much faster pace than anticipated, with monstrous social consequences.
1,086
3,052
661
24,189
2,595,641
Caliber retweeted
Tokens, training steps and accelerator-hours answer different planning questions. Our offline capacity software keeps each workload's units separate. A conversion needs evidence for that workload and hardware before it can support a delivery forecast.
1
1
4
Caliber retweeted
I have absolutely never ever "asked" to stop R&D on LLMs. In fact, I was very supportive of LLM efforts at Meta. Just look up my public advocacy for Galactica (November 2022), and the open sourcing of OPT-175B (May 2022), and of the Llama herd (mid 2023). I was actually influential in convincing the Meta leadership to start an AI-first product division built around Llama in January 2023. I've always said that LLMs were impressive, useful, and had amazing potential. I've also always said they were an off-ramp on the path towards human-level AI. Stop making me say things I didn't say or claim that I somehow hindered LLM R&D at Meta. That's BS.
51
41
3
787
79,765
This is the biggest challenge vis a vis the labs but recall that large companies have always had a huge resource advantage and yet sometimes small, highly motivated teams with incredible vision, focus, cohesion and insight blow away a big team. Welcome to capitalism 101.
If you are competing against OpenAI or Anthropic, you are going up against people using models two generations ahead of the best public models in the world, burning through $7k–$10k worth of tokens per day on an infinite token budget, running at 8x normal speed.
6
3
1
28
4,062
Caliber retweeted
Me running my orchestra of agents from the terminal:
57
176
24
1,489
123,739
Caliber retweeted
AI can polish a product memo, technical proposal, strategy brief, or essay. Can you defend the reasoning behind it? Caliber evaluates the evidence, trade-offs, and judgment behind AI-assisted work across professional and academic settings.

3
17
13,423
Today, we’re launching our home robot Isaac 1. Isaac 1 deliveries will begin this fall. Order yours below.
1,597
1,123
1,997
16,604
18,826,662
Enhancing Human Potential... For most of working life, polish was a decent stand-in for thought. If the analysis read well, someone had probably done the reasoning, and that rough correlation is what let managers and teachers and clients judge quality at a glance. AI broke the correlation. The analysis reads well now whether or not any reasoning happened behind it. The work is as good, or as thin, as it ever was. What's gone is your ability to tell which, from the page in front of you. And you're still making the call on a cue that has gone silent. caliber.dimaggi.ai/meter
1
1
23
There is a narrow window to embed structural security protocols before multi-agent systems scale globally. We believe this multilayered approach to agent security should be a collaborative priority for AI labs, government, and academia. See the framework → goo.gle/4vis97Q
9
21
4
109
33,561
Our data shows that the vast majority of issues don't stem from bad intent. They usually happen because an agent misinterprets a command or gets overly enthusiastic to achieve a goal. Understanding these nuances is critical for refining safety and security protocols. ⬇️
10
15
3
109
27,906
Instead of assuming AI will always do what we intend, we ask: what if it doesn't? That’s why we’ve developed our AI Control Roadmap: a framework for building and managing the advanced AI we deploy within Google. 🧵
89
90
33
593
93,610
Caliber retweeted
The true success of this approach hinges on ensuring local data sovereignty, open standards, and tangible local ownership in its partnerships, such as those with Anthropic.
1
1
61
Built on top of Gemini 3, Nano Banana Pro takes everything you loved and turns the dial to 💯: 🍌 SOTA Image Generation & Editing Sharper visuals, richer detail, and far more creative flexibility. 🍌 Smarter World Knowledge Understands objects, context, and relationships with engineer-level precision. 🍌 Accurate Text Rendering Generate infographics, diagrams, labels, and UI mockups with text that actually looks right. 🍌 Precision + Control Finer prompts. Better editing. More predictable outcomes. Nano Banana Pro isn’t just a model — it’s how engineers see the world: structured, detailed, and beautifully complex. If Nano Banana was fun, Nano Banana Pro is functional power — with flavor.
1
1
81
🍌 Banana lovers, brace yourselves… Nano Banana Pro just dropped. If Nano Banana was the snack, Nano Banana Pro is the full-course meal. And yes — it’s that good. Meet the model that sees the world like an engineer and creates like an artist: ⚡ Ultra-sharp image generation ⚡ Surgical-grade editing ⚡ Infographic wizardry ⚡ Text that actually looks like text ⚡ More brains. More control. More… banana. All powered by Gemini 3 for deeper knowledge, smarter reasoning, and visuals that hit different. Nano Banana Pro isn’t just an upgrade — it’s what happens when creativity goes pro mode. Go ahead. Peel back the future. 🍌✨
1
1
51
Meet Nano Banana Pro — smarter, sharper, and way more powerful. Better images, cleaner text, pro-level editing, and Gemini 3 intelligence packed into one 🍌
1
2
70