News from the AI & ML world

DeeperML - #ai

@www.microsoft.com //
Microsoft and Towards AI are addressing the AI skills gap with a new course, "AI for Business Professionals," designed to help professionals move beyond basic AI usage to achieve significant improvements in work quality and innovation. This initiative comes as organizations increasingly recognize generative AI as a game-changer but struggle with effective adoption due to a lack of strategic knowledge and technical skills among their teams. The course aims to transform individuals from merely AI-curious to AI-skilled collaborators, enabling them to use AI not just for speed, but to generate innovative ideas and achieve exceptional quality in their work.

The "AI for Business Professionals" course offers practical training tailored for diverse roles, including software engineers and investment research analysts. It provides modules and actionable strategies designed to optimize coding, streamline administrative tasks, and identify groundbreaking opportunities using AI. The self-paced course includes short videos, hands-on exercises, real-world demos, and expert guidance, all designed for busy, non-technical professionals. Participants will learn how to use AI deeply, effectively, and strategically in their daily work, addressing common issues such as distrust in AI output and uncertainty about time savings.

Microsoft emphasized the importance of continuous learning and career development in keeping up with evolving business needs. Nearly half (47%) of businesses say their top workforce strategy over the next 12 to 18 months is to train their existing workforce in AI skills. The course aims to provide individuals with the opportunity to develop the AI skills they need to build confidence, establish fluency, and thrive in the new AI economy. A free lesson is available to preview the course's content and learn how to use AI effectively, and strategically in daily work. The full course is available for $399.

Recommended read:
References :

@futurumgroup.com //
References: futurumgroup.com
NVIDIA is significantly advancing the field of humanoid robotics through its Isaac GR00T platform and Blackwell systems. These tools aim to accelerate the development and deployment of robots in manufacturing and other industries. Key to this advancement is NVIDIA's focus on simulation-first, AI-driven methodologies, leveraging synthetic data and integrated infrastructure to overcome the limitations of traditional robot training. This approach is particularly beneficial for European manufacturers seeking to enhance their production processes through AI.

NVIDIA's commitment to AI-powered robotics is evidenced by its substantial investment in hardware and software. The company is constructing the "world's first" industrial AI cloud in Germany, featuring 10,000 GPUs, DGX B200 systems, and RTX Pro servers. This infrastructure will support CUDA-X libraries, RTX, and Omniverse-accelerated workloads, providing a powerful platform for European manufacturers to develop and deploy AI-driven robots. NVIDIA's Isaac GR00T N1.5, an open foundation model for humanoid robot reasoning and skills, is now available, further empowering developers to create adaptable and instruction-following robots.

European robotics companies are already embracing NVIDIA's technologies. Companies such as Agile Robots, Humanoid, Neura Robotics, Universal Robots, Vorwerk and Wandelbots are showcasing AI-driven robots powered by NVIDIA's platform. NVIDIA is also releasing new models and tools, including NVIDIA Halos, a safety system designed to unify hardware, AI models, software, tools, and services, to promote safety across the entire development lifecycle of AI-driven robots. By addressing both the performance and safety aspects of robotics, NVIDIA is positioning itself as a key player in the future of AI-powered automation.

Recommended read:
References :
  • futurumgroup.com: Olivier Blanchard, Research Director at Futurum, shares insights on how NVIDIA’s Isaac GR00T platform and Blackwell systems aim to accelerate humanoid robotics through simulation, synthetic data, and integrated infrastructure.

Ruben Circelli@PCMag Middle East ai //
References: PCMag Middle East ai
Google is making significant strides in the realm of artificial intelligence with advancements in both video generation and browser assistance. The company's new Veo 3 AI video generator is capable of creating realistic videos from simple text prompts, marking a potentially revolutionary step in generative AI technology. Furthermore, Google is expanding access to Project Mariner, its AI-driven browser assistant, to a wider audience of Ultra plan subscribers, bringing more advanced features to users seeking enhanced web navigation and automation. These developments highlight Google's continued investment in and exploration of AI-powered tools designed to improve productivity and user experience.

The introduction of Veo 3 has sparked both excitement and concern. While the technology is undeniably impressive, with the ability to render finely detailed objects and create realistic audio, it also raises serious questions about the future of authenticity online. The potential for misuse, including the creation of deepfakes, online harassment, and the spread of misinformation, is significant. Experts worry that combining Veo 3's capabilities with weak content restrictions could lead to a catastrophic erosion of truth on the internet, especially once the ability to upload images for video generation is added. The implications of easily creating lifelike videos of individuals saying or doing things they never would are profound and potentially damaging.

In other AI developments, Google is rolling out Project Mariner to more Ultra plan subscribers, positioning it as a browser agent that interacts with open Chrome tabs via a dedicated extension. This allows Mariner to query and manipulate information from webpages, similar to other agent browsers. Users can instruct Mariner through a prompt bar, enabling tasks such as web navigation, hotel booking, and automated searches. However, the tool's frequent permission requests have led to feedback that it can be slow and requires significant manual oversight, limiting its autonomous value. While Google sees Project Mariner as a long-term bet within its AI-powered productivity suite, the immediate benefits may be overshadowed by its limitations.

Recommended read:
References :
  • PCMag Middle East ai: Combining instant photorealistic videos with Google's weak content restrictions is more than a recipe for disaster. It could mean the end of authenticity online forever.

@www.eweek.com //
References: www.eweek.com , Quartz , the-decoder.com ...
Meta is making a significant push into AI superintelligence, launching a new AI lab spearheaded by CEO Mark Zuckerberg. The move signifies a strategic shift, with Zuckerberg taking a more hands-on approach, operating in what insiders describe as "founder mode." This intensified focus aims to propel Meta ahead of competitors in the development of Artificial General Intelligence (AGI), striving to create AI systems capable of surpassing human intelligence. The new lab is part of a broader reorganization of Meta's AI efforts, reflecting Zuckerberg's determination to achieve leadership in the rapidly evolving AI landscape.

Meta's investment in achieving superintelligence involves significant financial commitments and strategic partnerships. Reports indicate that Meta is in talks to invest over $10 billion in Scale AI, a data platform company crucial for training AI models. Scale AI's CEO, Alexandr Wang, is expected to play a key role in the initiative, bringing expertise in large-scale AI infrastructure and data operations. This partnership will also see additional Scale AI staff transferring to Meta, marking a substantial external investment for the tech giant. Furthermore, Meta is reportedly offering lucrative compensation packages to attract top AI researchers and engineers, with some offers reaching seven to nine figures.

Zuckerberg's increased involvement in AI development stems from frustration with recent product releases, particularly the performance of Llama 4 and subsequent delays in launching its successor, Behemoth. These setbacks have motivated him to personally recruit and assemble an elite team of approximately 50 individuals dedicated to superintelligence research. He has reportedly created a WhatsApp group called "Recruiting Party" and is hosting private meetings to attract top AI talent. The ultimate goal is to integrate AGI across Meta's product suite, including chatbots and augmented reality applications, demonstrating the company's commitment to realizing the potential of superintelligent AI.

Recommended read:
References :
  • www.eweek.com: Meta’s Bold New Lab Takes Aim at Superintelligent AI, As Zuckerberg Enters ‘Founder Mode’
  • Quartz: Meta is making a huge push for AI 'superintelligence'
  • techstrong.ai: Meta Platforms Inc. is creating an artificial intelligence (AI) lab to pursue an AI system that surpasses human intelligence with "superintelligence."
  • the-decoder.com: Meta CEO Mark Zuckerberg is personally assembling a new team of experts to close the gap in artificial intelligence development.
  • www.unite.ai: Meta's reported $10 billion investment in Scale AI represents far more than a simple funding round—it signals a fundamental strategic evolution in how tech giants view the AI arms race.
  • Analytics India Magazine: Why Meta is Investing in Scale AI
  • eWEEK: Mark Zuckerberg enters ‘founder mode’ to personally build Meta’s new Superintelligence Lab, aiming to leap ahead in the AGI race with a handpicked team.
  • SiliconANGLE: Meta reportedly forming superintelligence lab amid Llama 4 Behemoth delays

Carl Franzen@AI News | VentureBeat //
Mistral AI has launched its first reasoning model, Magistral, signaling a commitment to open-source AI development. The Magistral family features two models: Magistral Small, a 24-billion parameter model available with open weights under the Apache 2.0 license, and Magistral Medium, a proprietary model accessible through an API. This dual release strategy aims to cater to both enterprise clients seeking advanced reasoning capabilities and the broader AI community interested in open-source innovation.

Mistral's decision to release Magistral Small under the permissive Apache 2.0 license marks a significant return to its open-source roots. The license allows for the free use, modification, and distribution of the model's source code, even for commercial purposes. This empowers startups and established companies to build and deploy their own applications on top of Mistral’s latest reasoning architecture, without the burdens of licensing fees or vendor lock-in. The release serves as a powerful counter-narrative, reaffirming Mistral’s dedication to arming the open community with cutting-edge tools.

Magistral Medium demonstrates competitive performance in the reasoning arena, according to internal benchmarks released by Mistral. The model was tested against its predecessor, Mistral-Medium 3, and models from Deepseek. Furthermore, Mistral's Agents API's Handoffs feature facilitates smart, multi-agent workflows, allowing different agents to collaborate on complex tasks. This enables modular and efficient problem-solving, as demonstrated in systems where agents collaborate to answer inflation-related questions.

Recommended read:
References :
  • Simon Willison: Mistral's first reasoning LLM - Magistral - was released today and is available in two sizes, an open weights (Apache 2) 24B model called Magistral Small and an API/hosted only model called Magistral Medium. My notes here, including running Small locally with Ollama and accessing Medium via my llm-mistral plugin
  • Simon Willison's Weblog: Mistral's first reasoning model is out today, in two sizes. There's a 24B Apache 2 licensed open-weights model called Magistral Small (actually Magistral-Small-2506), and a larger API-only model called Magistral Medium.
  • THE DECODER: Mistral launches Europe's first reasoning model Magistral but lags behind competitors
  • AI News | VentureBeat: The company is signaling that the future of reasoning AI will be both powerful and, in a meaningful way, open to all.
  • www.marktechpost.com: How to Create Smart Multi-Agent Workflows Using the Mistral Agents API’s Handoffs Feature
  • TestingCatalog: Mistral AI debuts Magistral models focused on advanced reasoning
  • the-decoder.com: The French start-up Mistral is launching its first reasoning model on the market with Magistral. It is designed to enable logical thinking in European languages.
  • www.artificialintelligence-news.com: Mistral AI has pulled back the curtain on Magistral, their first model specifically built for reasoning tasks.
  • www.infoworld.com: Mistral AI unveils Magistral reasoning model
  • AI News: Mistral AI has pulled back the curtain on Magistral, their first model specifically built for reasoning tasks.
  • Simon Willison: Mistral's first reasoning LLM - Magistral - was released today and is available in two sizes, an open weights (Apache 2) 24B model called Magistral Small and an API/hosted only model called Magistral Medium.
  • SiliconANGLE: Mistral AI debuts new Magistral series of reasoning LLMs.
  • siliconangle.com: Mistral AI debuts new Magistral series of reasoning LLMs
  • MarkTechPost: Mistral AI Releases Magistral Series: Advanced Chain-of-Thought LLMs for Enterprise and Open-Source Applications
  • www.marktechpost.com: Mistral AI Releases Magistral Series: Advanced Chain-of-Thought LLMs for Enterprise and Open-Source Applications

@the-decoder.com //
References: bsky.app , the-decoder.com , Maginative ...
OpenAI has recently unveiled o3-pro, a new AI model aimed at providing enhanced reliability and tool use for enterprise applications. While the model sacrifices some speed, it focuses on delivering more accurate and dependable responses, catering to the needs of businesses and developers requiring high-quality AI performance. Alongside the launch of o3-pro, OpenAI has also significantly reduced the price of its o3 model by 80%, making advanced AI capabilities more accessible to a wider range of users.

The o3-pro model is designed to excel in complex tasks, leveraging tools such as web browsing, file analysis, Python code execution, and memory of previous conversations to provide detailed and reliable outputs. Expert evaluations have consistently favored o3-pro over the standard o3 model in critical areas like science, programming, and business applications, citing improvements in accuracy, clarity, and adherence to instructions. This positions o3-pro as a top-performing option for education, programming, business, and even writing assistance, though responses may take longer due to the integrated tools.

The substantial price reduction of the o3 model and the introduction of o3-pro mark a strategic move by OpenAI to democratize access to high-tier reasoning models. By lowering the cost barrier, OpenAI aims to encourage broader adoption and experimentation with its AI technologies among startups, research teams, and individual developers. The o3 model is now priced at $2 per million input tokens and $8 per million output tokens, reflecting OpenAI's commitment to making frontier-level AI performance more affordable and accessible to a diverse user base.

Recommended read:
References :
  • bsky.app: The OpenAI API is back to running at 100% again, plus we dropped o3 prices by 80% and launched o3-pro - enjoy!
  • the-decoder.com: OpenAI lowered the price of its o3 language model by 80 percent, CEO Sam Altman said.
  • AI News | VentureBeat: OpenAI released the latest in its o-series of reasoning model that promises more reliable and accurate responses for enterprises.
  • Maginative: OpenAI’s new o3-pro model is now available in ChatGPT and the API, offering top-tier performance in math, science, and coding—at a dramatically lower price.
  • THE DECODER: OpenAI has lowered the price of its o3 language model by 80 percent, CEO Sam Altman said. The article appeared first on The Decoder.
  • AI News | VentureBeat: OpenAI's most powerful reasoning model, o3, is now 80% cheaper, making it more affordable for businesses, researchers, and individual developers.
  • www.cnbc.com: The figure includes sales from the company’s consumer products, ChatGPT business products and its application programming interface, or API.
  • Latent.Space: OpenAI just dropped the price of their o3 model by 80% today and launched o3-pro.
  • Simon Willison's Weblog: OpenAI's Adam Groth explained that the engineers have optimized inference, allowing a significant price reduction for the o3 model.
  • Sam Altman: We are past the event horizon; the takeoff has started. Humanity is close to building digital superintelligence, and at least so far it’s much less weird than it seems like it should be.
  • siliconangle.com: OpenAI’s newest reasoning model o3-pro surpasses rivals on multiple benchmarks, but it’s not very fast
  • SiliconANGLE: Silicon Angle reports on OpenAI’s newest reasoning model o3-pro surpassing rivals.
  • bsky.app: the OpenAI API is back to running at 100% again, plus we dropped o3 prices by 80% and launched o3-pro - enjoy!
  • bsky.app: OpenAI has launched o3-pro. The new model is available to ChatGPT Pro and Team subscribers and in OpenAI’s API now, while Enterprise and Edu subscribers will get access next week. If you use reasoning models like o1 or o3, try o3-pro, which is much smarter and better at using external tools.

@techinformed.com //
NVIDIA CEO Jensen Huang and UK Prime Minister Keir Starmer have recently joined forces at London Tech Week to solidify the UK's position as a leading force in AI. This collaborative effort aims to bolster the nation's digital infrastructure and promote AI development across various sectors. Starmer has committed £1 billion in investment to supercharge the AI sector, emphasizing the UK's ambition to be at the forefront of AI innovation rather than simply consuming the technology. Huang highlighted the UK's rich AI community, world-class universities, and significant AI capital investment as key factors positioning it for success.

The partnership includes the establishment of a dedicated NVIDIA AI Technology Center in the UK. This center will provide hands-on training in crucial areas such as AI, data science, and accelerated computing, with a specific focus on nurturing talent in foundation model building, embodied AI, materials science, and earth systems modeling. The initiative aims to tackle the existing AI skills gap, ensuring that the UK has a workforce capable of leveraging the new infrastructure and technologies being developed. Cloud providers are also stepping up with significant GPU deployments, with Nscale planning 10,000 NVIDIA Blackwell GPUs by late 2026 and Nebius revealing plans for an AI factory with 4,000 NVIDIA Blackwell GPUs.

Furthermore, the UK's financial sector is set to benefit from this AI push. The Financial Conduct Authority (FCA) is launching a 'supercharged sandbox' scheme, allowing banks and other City firms to experiment safely with NVIDIA AI products. This initiative aims to speed up innovation and boost UK growth by integrating AI into the financial sector. Potential applications include intercepting authorized push payment fraud and identifying stock market manipulation, showcasing the potential of AI to enhance customer service and data analytics within the financial industry.

Recommended read:
References :
  • blogs.nvidia.com: UK Prime Minister, NVIDIA CEO Set the Stage as AI Lights Up Europe
  • www.artificialintelligence-news.com: UK tackles AI skills gap through NVIDIA partnership
  • techinformed.com: techinformed reports Nvidia can boost UK’s digital infrastructure, says Huang as Starmer promises £1bn for AI
  • www.theguardian.com: UK banks to experiment with Nvidia AI in ‘supercharged sandbox’ scheme
  • NVIDIA Newsroom: UK Prime Minister, NVIDIA CEO Set the Stage as AI Lights Up Europe
  • blogs.nvidia.com: U.K. Prime Minister Keir Starmer’s ambition for Britain to be an “AI maker, not an AI taker,” is becoming a reality at London Tech Week.
  • ComputerWeekly.com: Starmer opens London Tech Week with £1bn AI boost
  • NVIDIA Newsroom: ‘AI Maker, Not an AI Taker’: UK Builds Its Vision With NVIDIA Infrastructure
  • TechInformed: Nvidia can boost UK’s digital infrastructure, says Huang as Starmer promises £1bn for AI
  • insidehpc.com: LRZ to Acquire HPE-NVIDIA ‘Blue Lion’ Supercomputer

Michael Nuñez@AI News | VentureBeat //
Apple is making significant strides in the field of artificial intelligence, particularly with its new image generation technology. Apple's machine learning research team has developed STARFlow, a breakthrough AI system that rivals the performance of popular image generators like DALL-E and Midjourney. STARFlow combines normalizing flows with autoregressive transformers to achieve what the team calls “competitive performance” with state-of-the-art diffusion models. This advancement comes at a critical time for Apple, which has faced increasing criticism over its progress in artificial intelligence, showcasing its broader effort to develop distinctive AI capabilities that differentiate its products from competitors.

This research tackles the challenge of scaling normalizing flows to work effectively with high-resolution images, something that has traditionally been overshadowed by diffusion models and generative adversarial networks. The STARFlow system demonstrates versatility across different types of image synthesis challenges, achieving competitive performance in both class-conditional and text-conditional image generation tasks. The research team includes Apple machine learning researchers along with academic collaborators, highlighting the company's commitment to pushing the boundaries of AI image generation.

Despite the image generation advancements, Apple Intelligence took a backseat at WWDC 2025. While Apple is giving developers access to Apple's on-device large language model (LLM) and introducing features like Live Translation in Messages, FaceTime, and the Phone app, the excitement around Apple Intelligence was more muted compared to previous years. Craig Federighi, Apple's SVP of Software Engineering, indicated that Siri needs "more time to reach a high-quality bar," suggesting that significant AI upgrades to Siri are still under development.

Recommended read:
References :
  • AI News | VentureBeat: Apple makes major AI advance with image generation technology rivaling DALL-E and Midjourney
  • www.laptopmag.com: Apple isn’t just sharing its AI. It’s betting developers will finish the job.

Amanda Caswell@Latest from Tom's Guide //
Apple's Worldwide Developers Conference (WWDC) 2025 highlighted the continued development of Apple Intelligence, despite initial delays and underwhelming features from the previous year. While the spotlight shifted towards software revamps and new apps, Apple reaffirmed its commitment to AI by unveiling a series of enhancements and integrations across its ecosystem. Notably, the company emphasized the progression of Apple Intelligence with more capable and efficient models, teasing additional features to be revealed throughout the presentation.

Apple is expanding Apple Intelligence through access to its on-device foundation model to third-party developers, allowing them to implement offline AI features. These AI features will be private and come without API fees. Users gain deeper access through new Shortcuts actions that offer direct access to Apple Intelligence models. The AI action will also include the option to use ChatGPT instead.

A key update is the introduction of Live Translation, integrated into Messages, FaceTime, and the Phone app. This feature facilitates real-time language translation, automatically translating texts and displaying captions during conversations. Visual intelligence, will allow users to select an object and search for similar products. These enhancements demonstrate Apple's focus on providing practical and user-friendly AI tools across its devices, aiming to streamline communication and improve user experience.

Recommended read:
References :
  • PCMag Middle East ai: Apple Intelligence Takes a Backseat at WWDC 2025
  • THE DECODER: Here's every Apple Intelligence update Apple announced at WWDC 25
  • MacStories: Apple Intelligence Expands: Onscreen Visual Intelligence, Shortcuts, Third-Party Apps, and More
  • www.techradar.com: Apple Intelligence was firmly in the background at WWDC 2025 as iPad finally had its chance to shine
  • www.tomsguide.com: Everyone’s talking about 'Liquid Glass' — but these 5 WWDC 2025 AI features impressed me most
  • www.techradar.com: Apple Intelligence is a year old - here are 3 genuinely useful AI tools you should use on your Apple products
  • www.techradar.com: TechRadar and Tom's Guide sat down with Apple's Craig Federighi and Greg Joswiak to talk about the company's latest plans for integrating Siri and Apple Intelligence.
  • www.eweek.com: Visual intelligence will work across more apps this fall, among other AI features announced at Apple’s Worldwide Developers Conference.
  • www.laptopmag.com: Apple isn’t just sharing its AI. It’s betting developers will finish the job.