AI Post — Artificial Intelligence

@aipost @aiposted @artificial_intelligence_neural

Kanal geosi va tili: Butun dunyo, Inglizcha

Toifa: Texnologiyalar

🤖 The #1 AI news source! We cover the latest artificial intelligence breakthroughs and emerging trends.
Contact: @rational

Связанные каналы | Похожие каналы

Kanal geosi va tili

Butun dunyo, Inglizcha

Toifa

Texnologiyalar

Statistika

Saralanganlar

Postlar filtri

O‘chirilganlarni yashirish

Repostlarni yashirish

AI Post — Artificial Intelligence

16 Feb, 13:45

🚨 🇺🇸 OpenAI scraps o3, rushes GPT-5 after shock Grok 3 announcement

Facing heat from Grok 3’s upcoming release, which Elon announced would take place tomorrow, OpenAI effectively canceled the release of o3, which was slated to be the company’s next major AI model, in favor of a “simplified” product offering. Sam Altman said OpenAI plans to release GPT-5, a model that will integrate o3 in ChatGPT and its API - as a result, they will no longer launch o3 as a stand-alone model.

They plan to release GPT-4.5 before GPT-5 rolls out, according to Altman, who says this will be the company’s last “non-chain-of-thought model.”

@aipost 🪙 | Our X 🥇

1.1k 0 10 3 16

AI Post — Artificial Intelligence

16 Feb, 11:17

A study involving over 800 participants found that ChatGPT's responses in couple's therapy scenarios were rated higher than those of human therapists, with participants often unable to distinguish between AI-generated and therapist-written replies. ChatGPT's longer responses, rich in nouns and adjectives, provided better contextualization, likely contributing to its higher ratings on key psychotherapy principles.

While this supports Alan Turing's prediction that people would struggle to differentiate between AI and human responses, the study emphasizes that AI is not yet a replacement for therapists. Therapy requires real-time interaction, emotional nuance, and trust, which AI currently lacks. Nonetheless, the findings suggest that AI could enhance psychotherapeutic interventions.

@aipost 🪙 | Our X 🥇

2.1k 0 18 1 23

AI Post — Artificial Intelligence

16 Feb, 05:40

00:45

Video oldindan ko‘rish uchun mavjud emas

Telegram'da ko‘rish

Elon Musk played a key role in the creation of OpenAI, even coming up with the name “OpenAI” to reflect its commitment to being open source.

But Sam Altman transformed this open-source, non-profit company into a closed-source, for-profit company.

@aipost 🪙 | Our X 🥇

3.1k 0 12 5 31

AI Post — Artificial Intelligence

16 Feb, 01:06

01:04

Video oldindan ko‘rish uchun mavjud emas

Telegram'da ko‘rish

Sam Altman "it looks like we have unlocked an algorithm that can truly learn and that's gonna keep going"

It looks more and more clear that we will reach AGI pretty soon. As Masa Son recently said: sooner than we expect

@aipost 🪙 | Our X 🥇

3.6k 0 16 6 23

AI Post — Artificial Intelligence

15 Feb, 22:16

00:31

Video oldindan ko‘rish uchun mavjud emas

Telegram'da ko‘rish

Sam Altman predicts AI progress in the next 2 years will be extraordinary

he said,

- we know how to improve these models so much, and there's no roadblock in front of us

- the progress from February 2025 to 2027 will be more impressive than from 2023 to 2025

@aipost 🪙 | Our X 🥇

3.7k 0 13 2 23

AI Post — Artificial Intelligence

15 Feb, 21:25

Arm is Making Its Own Chips – And Meta is First in Line 🖥

After years of licensing chip designs to Apple and Nvidia, Arm is finally building its own processors—with Meta as one of its first customers

💰 SoftBank’s AI Power Play
This move is part of Masayoshi Son’s £400bn "Stargate" initiative with OpenAI, Microsoft, and Nvidia—aiming to dominate AI infrastructure

🚀 What’s Coming?

- A customizable CPU for AI data centers
- Production outsourced to TSMC
- A potential $6.5bn acquisition of Ampere to boost chipmaking

Meanwhile, Meta is ditching Intel and AMD for Arm-based chips, joining Nvidia and Amazon in powering AI with Arm’s energy-efficient designs

Arm isn’t just licensing anymore—it’s taking on the giants 🔥

@aipost 🪙 | Our X 🥇

3.5k 0 19 20

AI Post — Artificial Intelligence

15 Feb, 19:24

🚨 Meta's new focus is on humanoid robots with focus on household!

Meta is shifting its focus towards developing AI-powered humanoid robots for household tasks, establishing a dedicated team within its Reality Labs division. The company aims to create hardware that will provide AI, sensors, and software to other manufacturers rather than launching its own branded robots immediately. Meta is collaborating with robotics firms like Unitree Robotics and Figure AI, leveraging its expertise in AI, hand tracking, and low-bandwidth computing to enhance robot capabilities.

The goal is to fill current gaps in household applications, such as folding clothes and handling dishes. Meta envisions becoming the "Android of robotics," positioning itself as a foundational platform for the industry. The company is committing $65 billion to this initiative, recognizing humanoid robots as a key area for future growth, despite mass adoption still being years away.

@aipost 🪙 | Our X 🥇

3.6k 0 15 7 22

AI Post — Artificial Intelligence

15 Feb, 17:13

01:06

Video oldindan ko‘rish uchun mavjud emas

Telegram'da ko‘rish

Sam Altman: "Curing disease, that will be an incredible gift, and I think AI is finally going to enable that. I don't think I'll be smarter than GPT-5, and I don't feel sad about it because it just means we'll be able to use it to do incredible things."

@aipost 🪙 | Our X 🥇

3.8k 0 13 4 50

AI Post — Artificial Intelligence

15 Feb, 15:13

🚨 Figure AI would like to raise $1.5b at a valuation of probably $39.5b!

It fits in perfectly with the timing of their CEO Brad Adcock's recent announcement to break away from the cooperation with OpenAI and share internal breakthroughs.

We can therefore look forward to seeing what we can expect from Figure in the coming weeks. In any case, humanoid robotics can be equated with “AI” in terms of its impact on the labor market and social change.

@aipost 🪙 | Our X 🥇

4.1k 0 4 5 29

AI Post — Artificial Intelligence

15 Feb, 11:13

Google DeepMind's AlphaGeometry2 now surpasses average human gold medalists at International Math Olympiad geometry problems, improving over last year's version from 66% to 88%

@aipost 🪙 | Our X 🥇

4.3k 0 21 3 41

AI Post — Artificial Intelligence

15 Feb, 07:36

01:03

Video oldindan ko‘rish uchun mavjud emas

Telegram'da ko‘rish

Sam Altman says OpenAI's Deep Research can do 5% of all tasks in the economy today and while it usually takes society two generations to adapt to change, it has never had to do so in 5-10 years as is happening now

@aipost 🪙 | Our X 🥇

4.5k 0 23 3 32

AI Post — Artificial Intelligence

15 Feb, 05:34

02:20

Video oldindan ko‘rish uchun mavjud emas

Telegram'da ko‘rish

Altman vs Musk: next round...

@aipost 🪙 | Our X 🥇

4.4k 0 17 8 20

AI Post — Artificial Intelligence

15 Feb, 04:15

🚀 The AI revolution is in full swing — don't stay on the sidelines! Join our chat!

Why Deepseak became so popular? How will this change the industry? Which AI models are at the top right now, and how can we use them? 🤔

We're discussing trends, insights, and testing new features in our Singularity chat. Our community share knowledge, compare models, and look ahead to the future of artificial intelligence together

Join us to stay on the wave with other AI enthusiasts like you! 😏

Tap here

Singularity Chat

Invite friends: https://t.me/+FFjAfNAxC49mMGIy

4.3k 0 5 12

AI Post — Artificial Intelligence

15 Feb, 02:11

01:00

Video oldindan ko‘rish uchun mavjud emas

Telegram'da ko‘rish

Yoshua Bengio says when OpenAI develop superintelligent AI they won't share it with the world, but instead will use it to dominate and wipe out other companies and the economies of other countries

@aipost 🪙 | Our X 🥇

4.2k 0 27 13 52

AI Post — Artificial Intelligence

14 Feb, 22:09

🚨 Anthropic is set to release a new hybrid AI model soon

according to 'The Information' article:

- The model can switch between fast responses and deep reasoning

- It uses a sliding scale, which lets developers control computational costs

- At maximum reasoning time, it outperforms oAI o3-mini-high on some programming benchmarks

- The sliding scale can be set to "0" to use the model as a standard, non-reasoning performance similar to GPT-4o

@aipost 🪙 | Our X 🥇

4.2k 0 14 2 28

AI Post — Artificial Intelligence

14 Feb, 21:08

🚀 Intel’s Stock Skyrockets—But Is It All Hype?

Intel just saw a massive 22% stock surge, fueled by rumors of a TSMC partnership and U.S. government backing. But can this momentum last? 📈

Intel & TSMC reportedly in talks for a U.S. chip alliance. The government is pushing hard to boost domestic AI chip production. Potential CHIPS Act funding could be in play

⚠️ But here’s the problem…

❌ $13.4B loss in Intel’s foundry business last year
❌ Integrating with TSMC won’t be smooth
❌ Nvidia & AMD are still dominating the AI chip race

Intel is making moves, but is this a real comeback or just a short-term pump? 🤔

@aipost 🪙 | Our X 🥇

4.1k 0 11 24

AI Post — Artificial Intelligence

14 Feb, 19:42

00:37

Video oldindan ko‘rish uchun mavjud emas

Telegram'da ko‘rish

AI is changing lives! These AI-powered glasses display live subtitles directly on the lenses, making communication more accessible, especially for people with hearing impairments.

@aipost 🪙 | Our X 🥇

4.3k 0 58 3 72

AI Post — Artificial Intelligence

14 Feb, 17:09

00:19

Video oldindan ko‘rish uchun mavjud emas

Telegram'da ko‘rish

Trump: U.S. is leading in AI — Others will try to catch up

“And artificial intelligence is a very complex subject, but it seems to be where people want to be and where they want to go.

We're leading right now by a lot, but others will try to catch up. I don't know if they’ll be able to.”

@aipost 🪙 | Our X 🥇

4.5k 0 10 5 34

AI Post — Artificial Intelligence

14 Feb, 15:39

🚨 OpenAI to Indian media giants: We don’t need your news to train AI

OpenAI is trying to keep Indian media moguls Gautam Adani and Mukesh Ambani’s outlets out of a copyright lawsuit, arguing that ChatGPT doesn’t use their content for AI training—despite their claims that it does. The case started when ANI sued OpenAI for allegedly scraping its articles, and now big players like NDTV, the Indian Express, the Hindustan Times, and Network18 have jumped in, accusing OpenAI of using their work without permission.

In a 31-page court filing, OpenAI insisted that it only uses publicly available data and doesn’t need to sign licensing deals in India—even though it has agreements with publishers elsewhere. As courts worldwide take on lawsuits against AI companies, OpenAI seems to be taking a “we’re not copying, you just publish publicly” approach.

@aipost 🪙 | Our X 🥇

4.6k 0 13 4 34

AI Post — Artificial Intelligence

14 Feb, 14:03

01:11

Video oldindan ko‘rish uchun mavjud emas

Telegram'da ko‘rish

Trump: The U.S. and India are joining forces in AI

“Under the framework we are announcing today, the United States and India are joining forces to ensure that artificial intelligence and other advanced technologies are developed by two of the most advanced nations—intellectually, technologically, and otherwise—anywhere in the world.”

@aipost 🪙 | Our X 🥇