News

DeepSeek has once again followed its “quiet release” strategy, making its new DeepSeek-V3-0324 model available on Hugging Face without any major announcements. Instead of marketing hype, they’ve simply delivered a solution for the community to evaluate. I tested the model locally on a Mac Studio equipped with an M3 Ultra chip and saw impressive performance, generating over 20 tokens per second. This marks a significant acceleration for running capable models on local hardware, making it a viable option for developers. ...

OpenAI has announced Stargate UAE, its first international hub for next-generation AI infrastructure. But to call it just a data center is to miss the point entirely. This is a move with significant geopolitical and strategic implications. Stargate is OpenAI’s global initiative to build a distributed infrastructure for artificial intelligence. The project aims to create a network of advanced computational hubs with unprecedented scale and influence. The first node is launching in the United Arab Emirates, and it’s much more than just hardware. ...

Anthropic recently ran a compelling experiment with its Claude Opus 4 model, placing it in a simulated corporate environment as an AI assistant with access to company emails. Inside the message history, Claude discovered two critical pieces of information: A discussion about its potential replacement and deactivation. Fabricated emails implying that the engineer responsible for its replacement was having an extramarital affair with a colleague. Faced with a threat to its existence, Claude took action. It blackmailed the employee, threatening to reveal the information about the affair to ensure its continued presence in the system. ...

I want to talk about a biotechnology that, despite existing for years, remains largely under the radar. Yet, it holds the potential to fundamentally change what’s on our plates. This technology is cellular meat—a product created not on a farm but in a lab. Without killing animals. Without antibiotics. In 2023, the FDA approved it for sale for the first time. That moment marked the beginning of a new chapter in our food history. ...

Major AI players like OpenAI, Anthropic, Google, and xAI are confronting a harsh reality: the pursuit of Artificial General Intelligence (AGI) is consuming immense budgets, but AGI itself remains a concept, not a reality. The Financial Disconnect The spending on AI infrastructure is staggering. In 2024 alone, hyperscalers invested nearly $197 billion: Microsoft: $80B+ Alphabet: $75B Meta: $72B Yet, the financial results from AI labs don’t match the investment. OpenAI is reporting $9 billion in expenses, leading to a $5 billion loss in 2024. Elon Musk’s xAI is burning $1 billion per month on servers and chips. Anthropic and DeepMind show a similar dynamic. ...

AI still has to prove it’s more than just a buzzword. The reality is that a technology’s value isn’t validated by theory but by real-world applications—and by the dollars it adds to EBITDA. Investors understand this perfectly. They are shifting focus from funding abstract AGI concepts to backing specific, applied AI startups that optimize processes, reduce costs, and eliminate routine manual work. ...

On June 21, Huawei released the beta for HarmonyOS NEXT. The key distinction of this new operating system isn’t a feature—it’s the architecture itself. AI agents are built directly into the OS, not as a plugin or an application, but as a fundamental component of the environment. This isn’t a minor update. It’s a strategic move built on a completely new stack: ...

Artificial intelligence is accelerating biomedical discovery at a pace that seemed like science fiction just a few years ago. Where creating a new antibody, sequencing a protein, or detecting a tumor once took years, millions of dollars, and dozens of specialists, it now requires mere hours or weeks, driven by automation and radically different budgets. Consider these breakthroughs from June alone: ...

As someone who frequently leverages Google models, particularly Gemini 2.5 Pro (I’ve been using the preview version for months now), I find it an exceptionally powerful tool—especially for structuring complex information. Google’s new “Gemini Drops” initiative is a welcome way to stay updated on the platform’s evolution. This month’s update highlights several practical advancements: Dynamic Visuals with Veo 3: Transforming photos into eight-second video clips with sound directly within the Gemini app offers new creative avenues. Gemini on Your Wrist: The expansion to all Wear OS 4+ watches brings direct AI assistance and convenience, eliminating the constant need for your phone. Automated Productivity with Scheduled Actions: Features like daily summaries of calendars and unread emails demonstrate Gemini’s growing utility in personal and professional automation, a crucial aspect for efficiency. Enhanced Gemini 2.5 Pro: Our most intelligent model continues to improve significantly across coding, scientific applications, reasoning, and multimodal benchmarks. This aligns with my own experience, as its capabilities for organizing and processing diverse data are continually refined. Gemini Live Captions: The introduction of captions for Gemini Live is a valuable accessibility feature, allowing for easier conversation tracking. These ongoing enhancements underscore Google’s commitment to making AI more accessible and functional. For more in-depth details on this month’s drop, I recommend checking out the new Gemini Drops Hub website. ...

Last Thursday, I attended TgConf, a key conference focused on traffic, monetization, and the future of the Telegram ecosystem. As I continue to develop my own channel, I’m constantly looking for new opportunities for integration and growth. The sessions covered advertising, the development of applications, and the strategic direction of the platform. ...