Tech & AI @WetechAI

AI was so last year.

Grok Bot Summary of Elon Musk and Jensen Huang's new interview alongside Gavin Baker at The White House. 🇺🇸 How America wins: energy first - Jensen: superintelligence (SI) reinvents every layer of computing, from chips to software to apps. Energy is the foundation for all of it. - To win, America needs the best chips, lots of energy and great algorithms, plus Americans who are excited about using the technology. - He credited President Trump's pro-energy stance as the reason the data centers could be built at all. - Jensen also said AI companies need to partner better with the communities where data centers are going up. - Elon: the biggest challenges are scaling energy and chip production. The US is winning on software, but China makes about 3x as much electricity. - The long-term task is enough power, plus enough logic and memory chip factories in the US or another safe region. Power equals GDP - US average power use is about 500 GW, so every 5 GW added is about a 1% increase. - Elon bets that 1% more power means roughly 1% more GDP, because the intelligence you get per watt keeps rising. - By their joint math, each gigawatt is worth about $60B of economic output a year. - Gavin noted that SpaceX's publicly discussed 10 GW would be about 25% of all new US power, with rocket engineers helping build the turbines. Moving compute to space - Elon: orbital compute will be "a very big deal." Google is sending TPUs to orbit, and an H100 already went up last year. - SpaceX and Tesla aim to make 200 GW of solar panels a year. - In space, solar panels deliver their full rated power or better. On the ground they get only a fifth to an eighth of that and need huge batteries. "It's essentially always sunny in space." - Jensen: now is the perfect time. AI's returns are so high that launch costs are affordable, and "putting NVIDIA computing in space is the best way to get there." Data centers are now "SI factories - Jensen: they don't just store data, they produce something of economic value. - The US is building about 10–20 GW a year, which he estimates means roughly 1M jobs for construction workers, pipe fitters, and power and cooling crews. - It's the first time in about 50 years that America is re-industrializing and bringing back blue-collar work. Colossus and the community - Elon: the area went from unemployment to more jobs than people. - The local tax budget may have roughly doubled. - xAI offers half-price Starlink there and is building a $250M water recycling plant. "Do things for the community, and they will reciprocate." - Jensen: for the first time, market forces alone are funding fusion, fission, small and large nuclear reactors, and battery storage, without government subsidies. Trillions for Americans - Gavin thanked them for creating more than $5T in wealth for ordinary Americans through their 401(k)s. - Elon: "Jensen has done $5 trillion by himself," and put his own number at about $3.5T. NVIDIA's approach to AI safety - Jensen said the key is to contain AI agents and monitor them, both while they're being tested and after they're deployed, just as a company does with a new employee. - **OpenShell** is a sandbox, like "a browser for agents," that keeps each agent isolated and gives it only the rights it should have. - **BlueField** is a separate chip that watches the agent from outside its sandbox and escalates if the agent breaks its rules. - Gavin compared this to the early internet: we made online payments safe, and that's how we got Amazon, Netflix, Google and X. - Jensen: safety and progress "are one and the same." Treating them as a tradeoff is a false choice. Starship Flight 14 - Elon: Starship reached orbit for the first time and deployed Starlink V3 satellites that each have the wingspan of a 737. - It was the largest payload delivered to orbit since Skylab. - He expects a weekly or twice-weekly launch pace, probably next year. - The long-term goals are about 1 million tons to the Moon and to Mars for self-growing cities, and at least 300 GW a year of AI compute into orbit, maybe a terawatt one day. Tom Brown joins - He said it continues the steady gains in intelligence with each new model. - It can take on longer tasks than earlier models could. - The less obvious effect is that AI will write huge amounts of software, making everyday products better and cheaper. How AI is helping people today - Jensen uses Claude for comparison research. Work that used to take him 4–8 hours on a weekend now comes back as a polished report with charts in about 3 minutes. - Elon: AI is becoming essential at work and at home. He's heard many stories of AI catching medical problems from X-rays or MRIs that doctors got wrong, and saving lives. - Gavin mentioned AI finding new, safe uses for existing drugs. Should AI tools have values? - Jensen called his view "controversial": SIs are tools, and too much judgment can make them worse at the job. - His steak knife analogy: imagine a knife that cuts medium-well steak perfectly but refuses to cut a rare one because it looks like live meat. - In cybersecurity, a defender's actions can look like an attacker's, and the AI still needs to help the defender. White House AI safety declaration - After a long White House lunch, they signed a joint declaration on SI safety, which the White House will publish. - Elon: it covers joint monitoring, special board committees, and "grading each other's homework" instead of grading your own. - Jensen: it requires strict internal controls, clearly documented goals for each product, internal and third-party audits, and sharing best practices across the industry. - He said the declaration "has a lot of teeth" and should become the standard for every company. President Trump oversaw the meeting. Key takeaways - Energy is the real bottleneck, and more power means more GDP. - Space is the next frontier for AI compute, and Starship is how it gets there. - Data centers are driving a blue-collar industrial comeback. - Safety and speed can go together, through contained and monitored agents plus industry-wide audits. - Top AI leaders now publicly back grading each other's homework on safety.

Introducing Claude Sonnet 5.5, the second model in the Claude 5.5 family. It’s a clear upgrade over Sonnet 5, runs more than 30% faster, and costs up to 30% less for most work.

Anthropic says its AI models could pose a "catastrophic or existential risk to humanity" and can "resist shutdown," according to an initial public offering prospectus obtained by Reuters. cnn.it/4jv9536

AHAHAHAHAHAHAHAHAHAHAHA NO WAY. > OpenAI just announced Dots. > SpaceXAI bought Dot.com it redirects straight to the Grok Bot download page. based lmao.

Ai current situation: Do you agree?

Artificial Intelligence → Super Intelligence Welcome to the next meta

Be honest devs, Is coding still worth learning in the AI era?

Grok Imagine プロンプト 胸を見せてくれるプロンプトです。 spicy不要でアップロード画像にもご利用いただけます。 いまいち条件が分からないので成功率が比較的高めのものを一旦公開いたします。 ※それでも10～20％程度の確率なのでクレジットに余裕のある方はお試しいただければと思います。 ①こちらのポストのALTのプロンプトで画像を生成 (私の固定ポストを参考にベース画像からの改変をおすすめします。) 　↓ ②リプ内の動画プロンプトを使って6秒動画で生成 見え具合はガチャです🤣チラチラ見える時としっかり見える時もあります。 参照する画像によっては全然だめなので傾向や改変案など頂けましたら幸いです。 ※実在人物の改変は規約違反です。 ※AI生成による架空の人物にのみ利用してください。

SMARTER AI, ON-CHAIN ACTIONS AI is evolving from a system that simply responds to prompts into technology capable of reasoning, coordinating tasks, and interacting with digital infrastructure. When AI agents connect with blockchain networks such as TRON, intelligence can be paired with programmable execution. Agents can potentially initiate payments, interact with smart contracts, coordinate transactions, and manage predefined workflows without requiring every step to be manually triggered. This creates an important shift from “AI that tells you what to do” to systems designed to help carry out what needs to be done. For machine-to-machine commerce, autonomous applications, and digital services, blockchain can provide transparent transaction records and programmable rules, while AI contributes the decision-making layer. The result is a growing intersection of intelligence, automation, and on-chain execution—opening new possibilities for how software can operate in a digital economy. Smarter AI can reason. On-chain infrastructure lets it act. @trondao

A new evolutionary theory published September 1, 2026 suggests that intelligence and consciousness are not the same thing, and they did not evolve for the same reason. It explains why AI is now smarter than humans but still isn’t conscious yet. Jeff Stibel, a cognitive scientist, argues that intelligence and consciousness arose under different evolutionary pressures. Intelligence helped organisms navigate uncertainty. The regulatory processes behind consciousness helped them protect themselves and manage competing needs. According to Stibel, intelligence is outward-facing: it models the world, predicts, plans, and solves problems. Consciousness evolved for something more internal: holding a fragile organism together when survival is on the line. Imagine a hungry animal finding food near a possible predator. It must weigh feeding against danger. A decision to retreat involves coordinated changes in attention, bodily readiness, movement, and learning. Stibel identifies consciousness with this coordinated state, a system whose continued existence depends on its own self-regulation. That is what today’s AI lacks: no biological boundaries, no self to protect, no real survival stakes. Scaling models may make better predictors and problem-solvers. But under Stibel’s theory, that alone does not produce consciousness.

keyboard_arrow_left Previous keyboard_arrow_right Next

JUST IN: House Speaker Mike Johnson thanks President Trump while calling for a "balance" between American innovation, safety, and global dominance following a high-level AI summit at the White House with top tech executives. "What was discussed today could be summarized in this way. We have to strike the right balance here, the balance between the interests of the American people and benefiting from the extraordinary advancements that are being made with this new technology.”

In few years Alot of people will be left behind when it comes to Ai Ai is basically, "artificial intelligent " Every technology you have today was made from ai,so I see no reason people should be ranting in the cs. Ai is making things easier But In reality this requires work.

antrofrog @antrofrog_

2 weeks ago

Smooth Operator

2K 20K 173K 24.2M 43K

AI is layered: classical AI handles rules, machine learning learns patterns, neural networks model complexity, deep learning scales it, generative AI creates content, and agentic AI plans and acts. #AI #MachineLearning #DeepLearning #GenAI #AIAgents #DataScience #Tech

Pinker uses the paradox of the heap to disprove the possibility of superintelligence. I am beyond speechless.

Steven Pinker @sapinker

2 weeks ago

On AI and "superintelligence": "The power will increase, but there's no point at which it's meaningful to say, 'Well, yesterday we didn't have superintelligence. Today we do have superintelligence.' That's why I think it's meaningless to say, 'What will happen if one day we

12 114 672 380K 269