AI Post β€” Artificial Intelligence
1.13M subscribers
1.11K photos
1K videos
2.52K links
πŸ€– The #1 AI news source! We cover the latest artificial intelligence breakthroughs and emerging trends.

Contact: @rational
Download Telegram
AI Post β€” Artificial Intelligence
Nvidia released Parakeet V2, a new open-source automatic speech recognition AI

β€” Transcribes an hour of audio in a second
β€” Top model on the Open ASR, beating ElevenLabs’ Scribe and OpenAI’s Whisper
β€” 6.05% Word Error Rate
β€” Available under CC-BY-4.0 license

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
Media is too big
VIEW IN TELEGRAM
Former Google CEO Eric Schmidt-backed FutureHouse launched four 'superhuman' AI agents for scientific discovery

β€” Crow for general research
β€” Falcon for deep literature reviews
β€” Owl for identifying previous research
β€” Phoenix for chemistry workflows

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
Media is too big
VIEW IN TELEGRAM
Jensen Huang says AI creates a new software development job by fundamentally changing how software is built and run. The next wave? AI factories. "A single 1-gigawatt AI factory could generate $60 billion annually"

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
πŸ€– AI tech gets a boost: Engineers teach AI with light

Engineers have just created a chip that teaches AI using light instead of electricity and it could make today’s computers resemble dial-up modems. This innovative chip reshapes beams of light in real time to solve intricate problems, reducing power consumption and accelerating AI training without requiring the use of a single electron.

The chip demonstrated exceptional performance in standard AI tests, learning more rapidly with fewer components and eliminating the need for bulky circuitry. Penn’s team believes this represents a significant advancement towards light-powered computers that possess the remarkable ability to learn at lightning speed.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
⚠️ What Happens When Teachers Are Replaced With AI? This School Is Finding Out

The Alpha School in Brownsville, Texas, is experimenting with an AI-driven educational model, allowing students to complete core lessons in just two hours each day. Staff, referred to as "guides," focus on facilitating learning rather than traditional teaching. Students engage in personalized, tech-based lessons and spend afternoons developing life skills.

This innovative approach has resulted in high test scores and increased student motivation. Amid growing interest in AI in education, the school plans to expand further in Texas and Florida, reflecting a shift towards technology-integrated learning environments.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
πŸ‡ΊπŸ‡Έ The Pentagon built an AI to price metals

Tired of China dictating the prices of rare metals, the U.S. has developed an AI system that accurately predicts the cost of materials like nickel and cobalt. Unlike market manipulations, this AI relies solely on mathematical calculations. Now, it’s under the stewardship of a nonprofit organization called CMF, where car manufacturers, miners, and defense companies are utilizing it to secure better deals and minimize the risk of being exploited.

Trained on extensive confidential data, this AI possesses the ability to forecast future supply disruptions and even play a crucial role in determining the viability of constructing a U.S. nickel mine. Consequently, the U.S. government has created an AI that enhances the intelligence of its purchasing decisions.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
πŸ‡ΊπŸ‡Έ Google AI’s newest user? Your 8-year-old

Google is launching its Gemini AI chatbot for children under 13. What could possibly go wrong with allowing a literal child to converse with a robot that occasionally fabricates information? Parents received emails informing them that Gemini can assist with homework and storytelling. However, there’s also a concern that it might expose your child to inappropriate content. So, good luck with that.

On a positive note, Google assures us that it won’t use your child’s data to train its AI (this time). Welcome to the joys of parenting in 2025: monitor the chatbot, verify the answers, and pray for the best.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
This media is not supported in your browser
VIEW IN TELEGRAM
Michael Saylor says AI will buy Bitcoin.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
This media is not supported in your browser
VIEW IN TELEGRAM
OpenAI CPO, Kevin Weil:

It's not gonna be long before... we're gonna have a model that is the best competitive coder in the world.

And forever after that point, computers will be better than humans at writing code.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
⏺ Is Duolingo the face of an AI jobs crisis?

Duolingo has announced plans to replace contractors with AI, signaling a potential AI jobs crisis. Journalist Brian Merchant highlighted that the company has already cut around 10% of its contractor workforce in recent years, replacing roles in translation and writing with AI technology.

This trend reflects broader concerns about companies using AI to reduce labor costs and limit hiring, particularly impacting entry-level jobs and freelance creative positions. Merchant argues that this shift is driven by management decisions rather than a dystopian job apocalypse.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
⏺ Anthropic CEO Admits We Have No Idea How AI Works

Dario Amodei, CEO of Anthropic, recently acknowledged that the AI industry lacks a fundamental understanding of how artificial intelligence operates. In a personal essay, he outlined plans to develop an "MRI on AI" over the next decade to better understand AI systems and identify potential risks.

Amodei noted that this ignorance is unprecedented in technology history and highlighted concerns stemming from the rapid advancement of AI without a clear grasp of its inner workings. He emphasized the importance of understanding AI before it significantly impacts society and the economy. Anthropic aims to create safer AI while also investigating the interpretability of these systems.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
⏺ OpenAI agrees to buy Windsurf for about $3 billion

OpenAI has agreed to acquire the startup Windsurf for approximately $3 billion, according to reports by Bloomberg. This acquisition reflects OpenAI's ongoing efforts to expand its capabilities and strengthen its position in the AI market.

Windsurf specializes in advanced AI technologies, which could enhance OpenAI’s offerings and development initiatives. The deal underscores the growing trend of consolidation in the artificial intelligence sector as companies seek to innovate and compete effectively.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
😱 Actors Horrified as They Learn What Selling Their Faces as AI Actually Means

Actors are expressing horror after realizing the implications of selling their likenesses for use in AI-generated content. Many, like South Korean actor Simon Lee, discovered their images were used in misleading promotions, while British actor Connor Yeates found his face featured in political propaganda without consent.

Legal experts highlight that actors often don’t fully understand the contracts they sign, which can grant companies broad rights to their likeness. Despite recent moves by companies like Synthesia to involve actors in decision-making, protections remain limited, leaving many vulnerable to exploitation.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
OpenAI is considering Lifetime and Weekly subscriptions for ChatGPT - found in latest ChatGPT build!

This could be a game-changer for users who want more flexible pricing and access to top-tier AI.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
Media is too big
VIEW IN TELEGRAM
Mistral just made two big reveals:

β€” Medium 3, a multimodal AI matching or surpassing 3.7 Sonnet, GPT-4o, and Llama 4 Maverick across benchmarks with 8x lower costs

β€” Le Chat Enterprise with corporate tools like Google Drive, agent building, and more

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
⏺ Anthropic launches a program to support scientific research

Anthropic has launched an "AI for Science" program to support researchers in high-impact scientific projects, particularly in biology and life sciences. The initiative offers up to $20,000 in API credits to qualified researchers, enabling access to Anthropic’s AI models.

The program aims to assist in analyzing complex data, generating hypotheses, and enhancing research efficiency. Anthropic emphasizes the potential of AI to accelerate scientific discovery, despite some skepticism among researchers regarding AI's current reliability in guiding the scientific process. Applications will be reviewed monthly based on scientific merit and potential impact.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
⏺ Anysphere, which makes Cursor, has reportedly raised $900M at $9B valuation

Anysphere, the creator of the AI coding tool Cursor, has reportedly raised $900 million in a funding round led by Thrive Capital, with participation from Andreessen Horowitz (a16z) and Accel. This funding values the company at approximately $9 billion.

Cursor previously raised $100 million at a $2.6 billion valuation in December. The startup has attracted significant investor interest, reflecting a growing demand for AI-powered coding solutions. Other competitors in the space, like Windsurf, are also gaining attention from investors.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM
AI Post β€” Artificial Intelligence
πŸ‡ΊπŸ‡Έ 250+ CEOs to states: Teach AI in schools or Watch America fall behind

Over 250 CEOs, including those from Microsoft, Uber, Meta, Zoom, Airbnb, Dropbox, and more, have signed an open letter urging states to mandate AI and computer science as a graduation requirement. The reason behind this push is twofold: only 12 states currently require computer science as a graduation requirement, and only 6.4% of high school students are taking it. In contrast, countries like China and South Korea have made it mandatory.

The letter highlights the importance of computer science education beyond just learning; it addresses income disparities, job skills development, and maintaining U.S. competitiveness in AI. Students in computer science classes earn 8% more, with even greater benefits for underrepresented groups. The recent establishment of an AI education task force by President Trump underscores the urgency of teaching AI to children to prevent them from lagging in this rapidly evolving field.

@aipost πŸͺ™ | Our X πŸ₯‡
Please open Telegram to view this post
VIEW IN TELEGRAM