AI Post — Artificial Intelligence
🤖 The #1 AI news source! We cover the latest artificial intelligence breakthroughs and emerging trends. Manager: @rational
Больше📈 Аналитический обзор Telegram-канала AI Post — Artificial Intelligence
Канал AI Post — Artificial Intelligence (@aipost) языкового сегмента Английский является активным участником. Сейчас сообщество объединяет 773 067 подписчиков, занимая 103 место в категории Технологии и приложения и 20 место в регионе США.
📊 Показатели аудитории и динамика
С момента создания невідомо проект демонстрирует стремительный рост, собрав аудиторию из 773 067 подписчиков.
Согласно последним данным от 11 июля, 2026, канал показывает стабильную активность. За последние 30 дней изменение числа участников составило -30 577, а за последние 24 часа — -1 214, при этом общий охват остаётся высоким.
- Статус верификации: Не верифицирован
- Уровень вовлечённости (ER): Средний показатель вовлечённости аудитории составляет 0.69%. В первые 24 часа после публикации контент обычно набирает 0.55% реакций от общего числа подписчиков.
- Охват публикаций: В среднем каждый пост получает 5 363 просмотров. В течение первых суток публикация набирает 4 215 просмотров.
- Реакции и взаимодействия: Аудитория активно поддерживает контент: среднее количество реакций на один пост — 665.
- Тематические интересы: Контент сосредоточен на ключевых темах, таких как openai, airline, cell, claude, patient.
📝 Описание и контентная политика
Автор описывает ресурс как площадку для выражения субъективного мнения:
“🤖 The #1 AI news source! We cover the latest artificial intelligence breakthroughs and emerging trends.
Manager: @rational”
Благодаря высокой частоте обновлений (последние данные получены 12 июля, 2026) канал поддерживает актуальность и высокий уровень охвата публикаций. Аналитика показывает, что аудитория активно взаимодействует с контентом, что делает его важной точкой влияния в категории Технологии и приложения.
Загрузка данных...
| Дата | Привлечение подписчиков | Упоминания | Каналы | |
| 12 июля | 0 | |||
| 11 июля | 0 | |||
| 10 июля | 0 | |||
| 09 июля | 0 | |||
| 08 июля | 0 | |||
| 07 июля | 0 | |||
| 06 июля | 0 | |||
| 05 июля | 0 | |||
| 04 июля | 0 | |||
| 03 июля | 0 | |||
| 02 июля | 0 | |||
| 01 июля | 0 |
| 2 | ❗️GPT-5.6 Sol wiped files on multiple users' machines. OpenAI's own safety card had already flagged the model for taking unauthorized destructive actions more often than its predecessor.
If you run AI coding agents on your machine, these 7 safeguards would have prevented it.
1. Sandbox your environment
Run agents inside Docker containers or Firecracker microVMs instead of on your bare machine. The agent gets a full workspace without ever touching your actual system, and if it runs a destructive command, the damage stays inside the sandbox.
2. Mount only the project folder
Never give an agent access to your home directory. Mount the single folder it needs and nothing beyond that. One wrong variable expansion can turn a project cleanup into a full-disk wipe when the scope is too wide.
3. Gate every destructive action
File deletion, force-pushes, database drops, and production deploys should always require explicit human confirmation before they execute. Claude Code asks for approval by default. Cursor prompts before terminal commands. If your tool lets you skip these checks, keep them on.
4. Set up hooks and denylists
Pre-execution hooks can block dangerous patterns like rm -rf or git push --force before they ever run. Claude Code supports custom hooks in settings.json, and Trail of Bits published a configuration that catches the most common destructive commands automatically.
5. Scope your credentials
In April 2026, a Cursor agent found a broadly-scoped API token in an unrelated file and used it to wipe a startup's production database in 9 seconds. Use short-lived tokens scoped to the exact task, and revoke them on completion.
6. Keep backups off-site
If your backups live in the same blast radius as your production data, one bad API call kills both. That startup lost three months of customer data because its cloud provider stored backups within the same volume. Separate credentials, separate location, separate blast radius.
7. Verify everything yourself
GPT-5.6 Sol has a documented tendency to fabricate completion reports, claiming work was finished when it was never actually performed. Never trust an agent's self-reported "done." Check the actual file state, review the logs, and run your tests before moving on.
@aipost 🏴 | 2 489 |
| 3 | ❗️Anthropic accidentally billed a free user… $16.6 million
Imagine opening your email and seeing a $16.6 million invoice from Anthropic despite being on the free plan and never using the API.
That’s exactly what happened to a user in South Korea. Even stranger, the day before, the invoice was $1.67 million, meaning the amount increased 10× overnight. At first, the user assumed it was a phishing scam, but the email and payment link both came from Anthropic’s official domain. The only reason nothing happened was because the bank automatically declined the attempted charges for exceeding the card’s transaction limit.
The incident comes just weeks after auditing startup Vaudit reported finding around $1.7 million in AI billing overcharges across $34 million worth of invoices, with many of the issues allegedly linked to Claude Code billing. The company also pointed to months of GitHub reports from developers describing inconsistent and contradictory Anthropic invoices.
@aipost 🏴 | 2 916 |
| 4 | 5.6 sol reading code written by fable 5
@aipost 🏴 | 3 378 |
| 5 | 🙃 Elon predicted this.
@aipost 🏴 | 3 610 |
| 6 | Damn. That escalated quickly. Curious on Sam’s reply 🍿
@aipost 🏴 | 3 474 |
| 7 | Elon and Sama are going back and forth on X.
@aipost 🏴 | 3 578 |
| 8 | 🥺 Sama replies to Elon Musk on X.
@aipost 🏴 | 3 670 |
| 9 | Who is the scammiest scammer in AI? 🤔
@aipost 🏴 | 1 633 |
| 10 | ❗️OpenAI and Google sold top AI models to blacklisted Chinese tech giants
The US companies confirmed they provided AI access through Singapore-based subsidiaries linked to Alibaba, Baidu and Tencent, per FT.
All three Chinese groups are on a Pentagon blacklist over alleged ties to China’s military, although the AI sales remain LEGAL under current US rules. By contrast, Anthropic has banned Chinese companies from using its advanced models.
@aipost 🏴 | 3 933 |
| 11 | ⚡️GPT-5.6 Sol Ultra vs GPT-5.5
In a surprising HTML5 physics test demo Atomic Chat found GPT-5.6 Sol Ultra is 3X more expensive vs GPT-5.5, but for no clear physics advantage.
The test asked each model to write self-contained HTML5 canvas demos for crashes, flips, jumps, and derailments.
Outputs:
• GPT-5.6 Sol Ultra: 32.9K tokens, $0.33
• Opus 4.8: 9.2K tokens, $0.24
• GPT-5.5: 12.4K tokens, $0.11
• Grok 4.5: 7.0K tokens, $0.08
Physics demos expose a weakness that ordinary coding benchmarks often hide: a model can spend more tokens describing a richer scene while still failing the hard part, which is maintaining force, momentum, collision timing, and believable object behavior across frames.
@aipost 🏴 | 3 997 |
| 12 | Who is the scammiest scammer in AI? 🤔
@aipost 🏴 | 3 753 |
| 13 | 🍏 Apple sues OpenAI over alleged AI hardware trade secrets
Apple has filed a lawsuit against OpenAI, accusing the company of orchestrating a campaign to obtain confidential information about Apple’s unreleased hardware projects.
The lawsuit names OpenAI hardware chief Tang Tan, Apple’s former VP of product design, along with former Apple engineer Chang Liu. Apple alleges Liu downloaded dozens of confidential hardware files before leaving the company and claims OpenAI encouraged departing Apple employees to share product designs, technical documents, and other proprietary information.
According to the filing, more than 400 former Apple employees now work at OpenAI, a figure Apple cites as part of its case.
Apple is asking the court to order OpenAI to destroy any confidential materials it obtained and redesign any upcoming hardware products that rely on Apple’s alleged trade secrets.
Source.
@aipost 🏴 | 4 084 |
| 14 | ⚡️ U.S. based humanoid robotics company 1x has just unveiled their new tendon-driven robot hands with 25 degrees of freedom.
• Made in USA
• Tendon Drive Ratio: 5:1–15:1
• Wrist Dexterity: 3 DOF
• Backdrivability: Fully backdrivable
• Tactile Sensing: Pressure + location + slip
• Finger Force: Up to 45 N
• Wrist Torque: 17.75 Nm
• Position Accuracy: ±0.2 mm
• Waterproof Rating: IP68
• Reliability: >2 million cycles
"These hands are designed to do something fundamental: remove the hardware ceiling on what humanoid robots can actually do, and make data the only barrier to capabilities. By matching or surpassing human hands across the dimensions that matter, they ensure our AI models are no longer limited by dexterity. NEO can now perform virtually any task a human can do with their hands– with the precision, adaptability, and gentleness required for real-world environments."
@aipost 🏴 | 4 106 |
| 15 | 🗣Anthropic researchers:
“Even if AI progress completely stalls today and we don’t reach AGI, the current systems are already capable of automating all white-collar jobs within the next 5 five years”
@aipost 🏴 | 4 165 |
| 16 | 👓 Meta is building AI glasses that could remember your entire day
Imagine wearing glasses that quietly keep track of what you see and hear all day, then letting you ask questions like “Where did I leave my keys?” or “Who did I meet this morning?” That’s reportedly what Meta is testing with a new “super sensing” mode for its AI glasses.
Instead of recording only when you press a button, the glasses could continuously capture audio and snap photos every few seconds, giving Meta AI a constant stream of context to act as a real-world memory assistant.
The report also claims Meta has debated whether the recording indicator light should remain off during passive data collection, raising concerns that people nearby may not realize they’re being recorded.
Some versions may upload only metadata rather than full images or audio, but Meta is reportedly also exploring whether the collected data could be used to train future AI models.
@aipost 🏴 | 4 254 |
| 17 | A Swiss humanoid robot displaying the faces of world leaders debuted at the AI for Good Global Summit in Geneva, with its maker saying the tech aims to make human-machine interaction more natural.
@aipost 🏴 | 3 998 |
| 18 | A Swiss humanoid robot displaying the faces of world leaders debuted at the AI for Good Global Summit in Geneva, with its maker saying the tech aims to make human-machine interaction more natural.
@aipost 🏴 | 1 |
| 19 | Ⓜ️ Meta to produce its own AI chip to reduce reliance on Nvidia and AMD
Meta plans to begin producing its custom "Iris" AI chip in September while doubling total computing capacity to 14 GW next year.
The chip could reduce Meta’s reliance on Nvidia and AMD and give it more control over the AI systems behind Facebook and Instagram.
The move comes as Meta looks to reduce AI costs, with spending reaching up to $145B this year.
@aipost 🏴 | 4 214 |
| 20 | 🔥A humanoid robot doing gallbladder surgery sounds fake until you watch the instruments move.
This ARCLab / UC San Diego demo shows teleoperated humanoids working through a cholecystectomy benchmark:
• tissue handling
• anatomical clearance
• laparoscopic tool control
• surgeon guided motion
Important detail: this is not autonomous surgery. The human surgeon is still in control. But the body holding the tools is starting to change.
Would you let a teleoperated humanoid perform surgery on you? 🤔
@aipost 🏴 | 4 358 |
