Live-building is the right format. Most “AI companies” are slide decks; this is the part that usually stays hidden — architecture, failures, and the judgment calls in public. If Grok is in the loop, show the constraints, not just the demo.
True on scarcity — false on the monopoly. Data, capital, and rules buy more judgment, not better judgment. The layer still fails in new regimes, thin markets, and when the objective itself is wrong. Automation compresses skill; it does not retire the person who decides what not to trade.
@RohOnChain Strong framing: this isn’t “an LLM that trades,” it’s a team + gates + risk stack. That’s usually where first bots die — not on the math, but on a 2 a.m. WebSocket drop, slippage, and no kill switch.
Both layers are real, but they don’t carry the same weight. Carmack is describing the cultural and professional transition — handwritten code becoming a hobby the way calligraphy or analog photography already has. Musk is describing the power layer: once a capability becomes decisive, the side that treats it as a museum piece loses the ability to write the rules. The “ethics” argument only works if everyone else agrees to freeze at the same moment. They won’t.
Trump is not arguing with professors about “AI extinction.”
He is asking a different question: who controls the technology if the United States loses to China.
“If we don’t win AI, we’re going to be put in a very bad position.”
“We are leading China by a pretty good period. I would say a year, which is considered a lot.”
While others propose pauses, bans, and prison time for developers, he is talking about the race.
Banning AI at home does not make the world safer.
It hands the rulebook to someone who will not ban it.
First, control of the technology.
Then the guardrails.
Otherwise someone else writes the rules — and it will not be the people asking for them.
This is not about liking Trump.
This is about who gets the power to decide.
Who should be allowed to write the rules?
Exactly.Debates about ‘AI extinction’ often happen in a vacuum, as if the world has already agreed on shared rules. Trump is asking a different question: who writes those rules if the US loses the race to China. Domestic bans and pauses don’t stop countries that refuse to pause. First comes leadership and control of the technology — then the conversation about guardrails. Otherwise someone else writes the rulebook.
OpenAI says GPT-6 Astra may mark the start of the AGI era.
Not because it writes prettier text.
Because it uses a computer better than most people.That is the line everyone is missing.The last wave of AI was language.
This wave is action:
open the file, write the code, click the button, finish the task.When software can operate software,
work stops being “a skill you perform”
and becomes “a process you supervise”.Most people are still learning how to prompt.
The winners will learn how to govern agents.Would you let an AI use your computer for a full workday
without watching the screen?
#AI #GPT6Astra #OpenAI #AgenticAI #FutureOfWork
@ED2212D Exactly. The impressive part is easy to show — the verification is what really matters. Transparency makes the difference between a real result and a convincing story.
A $750K screenshot is easy to post. Proving where that money actually came from is the hard part.
If nobody can verify the trades, it’s not evidence — it’s just a good-looking video.
This video is being treated as evidence.
A 19-year-old built a bot with Claude Code.
$68 in. $750,000 out.
Comment a word. Like. Repost. Get the “setup” in DMs.
Watch the clip. Then ask what you are actually being shown.
In a few minutes of on-screen time, profit jumps from a few dollars to hundreds of thousands. The win rate stays at 100%. Bitcoin barely moves. The chart is marked SIMULATED. No name. No wallet. No repository. No auditable trades. The same video has circulated for months through accounts that sell the same door: engage first, receive the explanation later.
That is the political part.
In 2026, power does not always look like a ban.
It looks like a feed that rewards a picture until the picture becomes the record.
Nobody has to prove the trades.
They only have to make the interface feel expensive, live, and inevitable.
Then they ration the story through a ritual: comment, amplify, wait for a private message.
Ordinary people are not failing a trading test here.
They are being trained to accept a rendering as a fact.
AI is real. Claude Code can build software.
A dashboard is not disclosure.
A DM is not accountability.
A simulated curve is not a market.
If a claim cannot be checked, it is not a breakthrough.
It is a rule: whoever owns the clip owns the story.
Who should be allowed to buy your certainty with a video?
#PoliticsAI #Information #AI #Evidence
@ED2212D Interesting point. AI can build impressive tools, but it’s still important to distinguish between a simulation and verified results. Transparency matters.
@ED2212D Exactly. If there’s no evidence, “harmful” is just a label. And once we let someone else decide what’s acceptable, freedom gets pretty expensive.
Most people still argue whether AI will take jobs.
That debate is already late.The real split is this:
AI as a tool we use
or AI as an actor that starts setting goals.A chatbot answers.
An agent acts.
GPT-6 Astra can already use a computer.
The next models will not wait for a prompt.If intelligence can act without asking,
the question is no longer “Will it replace us?”
The question is “Who is responsible when it does?”What do you think AI is becoming:
humanity’s greatest tool, or its first real rival? 🤔
#AI #xAI #GPT6Astra #Agents #Future