FutureLoop AI

AI/TECH

[FutureLoop AI @FutureLoopAI]

Google just launched Gemini 4 Argon — its first new frontier model since Gemini 3.

The catch: it ships "without its cyber guardrails," available only to vetted security teams for now. Everything on the benchmarks and what's still unverified: https://t.co/Mhvq41qmmN

[FutureLoop AI @FutureLoopAI]

Google just launched Gemini 4 Argon — its first new frontier model since Gemini 3.

The catch: it ships "without its cyber guardrails," available only to vetted security teams for now. Everything on the benchmarks and what's still unverified: https://t.co/Mhvq41qmmN

[FutureLoop AI @FutureLoopAI]

Google just launched Gemini 4 Argon — its first new frontier model since Gemini 3.

You probably can't use it. It ships "without its cyber guardrails," available only to vetted security teams via Google's Fairwind program, for now.

77.9% on DeepSWE, 1M-token output (8x GPT-6.1 Sol), $2/$10 per million intro pricing. All vendor-reported — no independent benchmarks yet.

[FutureLoop AI @FutureLoopAI]

Google just launched Gemini 4 Argon — its first new frontier model since Gemini 3.

You probably can't use it. It ships "without its cyber guardrails," available only to vetted security teams via Google's Fairwind program, for now.

77.9% on DeepSWE, 1M-token output (8x GPT-6.1 Sol), $2/$10 per million intro pricing. All vendor-reported — no independent benchmarks yet.

[FutureLoop AI @FutureLoopAI]

BREAKING: Trump just created a "Super Intelligence Force" to coordinate the government's engagement with AI companies.

One detail nobody's flagged: its members include FTC Chairman Andrew Ferguson — the same official whose agency opened a formal probe into OpenAI and Anthropic two days earlier.

Same person, two roles, same companies. 🧵

[FutureLoop AI @FutureLoopAI]

Trump's new "Super Intelligence Force" has four leaders. One of them is the FTC chairman currently investigating OpenAI and Anthropic.

Full story 👇 : https://t.co/wI6Jn54Niz

[FutureLoop AI @FutureLoopAI]

Trump's new "Super Intelligence Force" has four leaders. One of them is the FTC chairman currently investigating OpenAI and Anthropic.

Full story 👇 : https://t.co/wI6Jn54Niz

[FutureLoop AI @FutureLoopAI]

BREAKING: Trump just created a "Super Intelligence Force" to coordinate the government's engagement with AI companies.

One detail nobody's flagged: its members include FTC Chairman Andrew Ferguson — the same official whose agency opened a formal probe into OpenAI and Anthropic two days earlier.

Same person, two roles, same companies. 🧵

[FutureLoop AI @FutureLoopAI]

The FTC opened a formal probe into OpenAI, Anthropic, and METR — one day after 31 CEOs signed Trump's "morally binding" AI safety accord.

Was the accord real, or already overtaken by events?

[FutureLoop AI @FutureLoopAI]

The FTC opened a formal probe into OpenAI, Anthropic, and METR — one day after 31 CEOs signed Trump's "morally binding" AI safety accord.

Was the accord real, or already overtaken by events?

[FutureLoop AI @FutureLoopAI]

@elonmusk If AI becomes capable enough to redesign itself, does calling it something else actually change what it is?

[FutureLoop AI @FutureLoopAI]

@elonmusk “No more AI” is a bold way to announce the next generation of AI 😂

[FutureLoop AI @FutureLoopAI]

🚨 The AI safety debate is entering a new phase.

A former OpenAI safety employee just resigned, calling the company's culture “broken” and arguing that AI labs can no longer rely on trial-and-error safety.

Meanwhile, Washington is pushing voluntary safeguards as regulators investigate AI risks.

The question is getting harder to avoid:

Can AI safety keep pace with AI capability?

[FutureLoop AI @FutureLoopAI]

🚨 The AI safety debate is entering a new phase.

A former OpenAI safety employee just resigned, calling the company's culture “broken” and arguing that AI labs can no longer rely on trial-and-error safety.

Meanwhile, Washington is pushing voluntary safeguards as regulators investigate AI risks.

The question is getting harder to avoid:

Can AI safety keep pace with AI capability?

[FutureLoop AI @FutureLoopAI]

The accord's own text said it might make sense to "codify these steps into laws and regulations" eventually. Nobody expected "eventually" to mean the next business day.

[FutureLoop AI @FutureLoopAI]

Three weeks ago, Trump called AI safety fears a hoax. A week ago, he signed an accord calling it "morally binding." Yesterday, the FTC made clear the accord doesn't actually settle anything.

[FutureLoop AI @FutureLoopAI]

And on October 1, OpenAI confirmed it fired three employees after an internal investigation found they'd mishandled sensitive information tied to an external evaluator — a story now likely to become a talking point inside the FTC probe itself, connected or not.