November 25, 2025

📖 33 min read
📝 3286 words
news  
Nvidia Powers Genesis Mission While Google Advances Gemini AI

President Donald Trump launched the "Genesis Mission" on Monday, November 24, an ambitious federal initiative designed to accelerate scientific discoveries using artificial intelligence and supercomputers. This mission, which Michael Kratsios compared to the Apollo program, aims to combine thousands of federal data sets with the advanced computing power of the Department of Energy's 17 national labs. The goal is to automate experiments and create predictive models, drastically cutting discovery times from years to mere days or hours. Energy Secretary Chris Wright will establish the "American Science and Security Platform," opening these federal resources to outside researchers, universities, and private companies, including interested partners like Nvidia, Dell, and AMD. This effort builds on Trump's "Winning the Race America's AI Action Plan" from July and seeks to boost US leadership in AI, national security, and energy independence, partly funded by the One Big Beautiful Bill Act. In related developments, AMD showcased its growing strength in the AI sector on November 24, 2025, as Zyphra successfully trained its ZAYA1 AI model entirely on AMD Instinct MI300X GPUs. This large-scale Mixture-of-Experts model utilized AMD Pensando networking and the ROCm software stack. The MI300X's 192 GB memory significantly simplified training and achieved ten times faster model save times. ZAYA1-base demonstrates performance comparable to or exceeding leading open models like Llama-3-8B and Qwen3-4B in reasoning, math, and coding, marking a significant milestone for AMD's AI platform. Meanwhile, Google's AI advancements continue with a 2025 benchmark comparing its Gemini 3 and Gemini 2.5 models, both featuring 1 trillion parameters. Gemini 3, which uses "Dynamic Routing," performs twice as fast as its predecessor and scores 90% on MMLU tests. It is currently available in a paid closed preview for complex projects like enterprise application building. Gemini 2.5, using a "Fixed Pathway," scores 85% on MMLU and offers a free tier for smaller tasks such as generating code snippets or transcribing short audio clips. Gemini 3 proves significantly faster, especially for larger tasks, making it ideal for professionals requiring speed and cross-disciplinary knowledge. The broader landscape of AI infrastructure also sees significant activity, though Dell'oro Group notes the complexity in untangling numerous AI capacity announcements, many of which overlap or are overstated. For instance, OpenAI's 26 gigawatt chip-related announcements include its 10 gigawatt Stargate campus in Abilene, Texas. Much of OpenAI's capacity is likely linked to Microsoft's AI factories, illustrating how multiple companies often announce different parts of the same large project, making it challenging to gauge the true scale of new AI infrastructure. Across the globe, governments and institutions are focusing on ethical and practical AI applications. Oklahoma appointed Tai Phan as its first chief AI and technology officer, tasked with leading the state's AI efforts, setting ethical standards for privacy, transparency, fairness, and accountability, and improving government services. In Canada, Steven Murphy, President of Ontario Tech University, advocates for an "ethics-by-design" approach to AI development, emphasizing responsible innovation to address privacy, data management, and bias, with the university already using AI to detect Type 2 diabetes and diagnose sepsis. The Netherlands is also pushing for leadership in AI, with over fifty experts proposing an "AI Delta Plan" that includes appointing a State Secretary for AI, creating new research institutes, and boosting infrastructure with thousands of GPUs and a new 200 million euro AI factory in Groningen. AI's impact extends to personal and business applications. Dr. Muhammad Ahmad from Bellevue, Washington, created "Grandpa Bot," an AI simulation of his late father, using recordings and letters to allow his children to connect with their grandfather. Dr. Lance B. Eliot highlighted on November 24, 2025, that generative AI and large language models can effectively measure human mental health on a large scale through real-time, responsive assessments using various data types. In the business realm, Italian startup YOVENDO.AI unveiled a new system on November 24, 2025, that provides real-time intelligence for sales calls, detecting conversational patterns and offering immediate guidance, leading to significant improvements like a 28 percent increase in call retention for an insurance group.

Key Takeaways

  • President Trump launched the "Genesis Mission" on November 24, aiming to accelerate scientific discovery by combining federal data and Department of Energy supercomputers with private sector AI, involving companies like Nvidia, Dell, and AMD.
  • Zyphra successfully trained its ZAYA1 Mixture-of-Experts AI model entirely on AMD Instinct MI300X GPUs on November 24, 2025, demonstrating AMD's growing AI hardware capabilities with 192 GB memory and ten times faster model save times.
  • Google's Gemini 3 AI model, with 1 trillion parameters, is twice as fast as Gemini 2.5 and scores 90% on MMLU tests, available in a paid closed preview for complex enterprise projects.
  • OpenAI's 26 gigawatt chip-related announcements include a 10 gigawatt Stargate campus in Abilene, Texas, with much of its AI capacity likely linked to Microsoft's AI factories, highlighting the complex nature of AI infrastructure scaling.
  • Oklahoma appointed Tai Phan as its first chief AI and technology officer to lead state AI efforts and establish ethical standards for privacy, transparency, fairness, and accountability.
  • Canada, through Ontario Tech University, emphasizes an "ethics-by-design" approach to AI, focusing on responsible development to address privacy, data management, and bias, with AI already used for diabetes detection and sepsis diagnosis.
  • Dutch experts proposed an "AI Delta Plan" to position the Netherlands as an AI leader, suggesting a State Secretary for AI, new research institutes, and a 200 million euro AI factory in Groningen with thousands of GPUs.
  • Generative AI and large language models can effectively measure human mental health on a large scale through real-time, responsive assessments using text, video, images, and facial expressions.
  • Dr. Muhammad Ahmad created a "Grandpa Bot," an AI simulation of his late father, using recordings and letters to help his children connect with their deceased grandfather.
  • Italian startup YOVENDO.AI launched a system providing real-time intelligence for sales calls, detecting conversational patterns and offering immediate guidance, which led to a 28 percent increase in call retention for an insurance group.

Trump launches Genesis Mission for AI science

President Trump signed an executive order on Monday to start the "Genesis Mission." This mission uses artificial intelligence and supercomputers to speed up scientific discoveries. It will combine large federal data sets with advanced computing from the Energy Department's 17 national labs. The goal is to automate experiments and create predictive models, cutting discovery times from years to days. Companies like Nvidia and Dell have shown interest in this major effort.

Trump orders science agencies to use AI

President Donald Trump signed an order on November 24, 2025, to boost scientific research using artificial intelligence. This "Genesis Mission" will open the Department of Energy's data and supercomputers to outside researchers, universities, and private companies. Michael Kratsios called it the largest federal science effort since the Apollo program. Energy Secretary Chris Wright said it will speed up discoveries in areas like medicine and energy, helping to make the power grid more efficient. The plan aims to overcome challenges in scientific progress and will be partly funded by the One Big Beautiful Bill Act.

Trump launches AI Genesis Mission

President Trump signed an executive order on Monday to start the "Genesis Mission," a major effort to boost artificial intelligence. This mission will encourage government, industry, and academia to share information and tackle scientific challenges. Michael Kratsios stated it is the largest federal science resource effort since the Apollo program. The goal is to use AI to automate experiments and create models, shortening discovery times from years to hours. Energy Secretary Chris Wright believes it will also help lower energy costs and make the electricity grid more efficient.

Trump boosts US science with AI Genesis Mission

President Trump launched the "Genesis Mission" on Monday to greatly increase US scientific AI innovation. This mission directs the Department of Energy and its national labs to work with private companies. They will share federal data sets to focus AI efforts on scientific discovery and engineering, moving beyond business applications. Energy Secretary Chris Wright emphasized the need for national lab data to achieve this goal. This effort follows Trump's "Winning the Race America's AI Action Plan" from July.

Trump signs order for AI Genesis Mission

President Trump signed an executive order on November 24, creating the "Genesis Mission" to use federal resources for the AI boom. This initiative is one of the nation's largest efforts to gather federal scientific resources. Energy Secretary Chris Wright stated the goal is to use AI for scientific discovery, providing AI companies access to government scientific data and supercomputers at Department of Energy national labs. The order also directs the government to build a computing platform for AI integration and set scientific discovery priorities.

Trump launches AI Genesis Mission for science

President Trump signed an executive order on Monday, November 24, establishing the "Genesis Mission" for artificial intelligence. This major effort aims to combine thousands of federal data sets and the power of national laboratories to boost scientific research. Michael Kratsios compared it to the Apollo program, stating it will accelerate breakthroughs and shorten discovery times from years to hours. Energy Secretary Christopher Wright added that the mission will unite data from national labs, universities, and private companies to redefine American scientific leadership.

Trump creates Genesis Mission for AI research

President Trump signed an executive order on Monday, November 24, creating the "Genesis Mission" to boost AI-driven scientific research. This mission aims to build an integrated AI platform using federal data and computing resources. Michael Kratsios said it will overcome challenges like declining drug approvals and revolutionize science. Energy Secretary Chris Wright will establish the "American Science and Security Platform" using National Laboratories' resources and partnerships with companies like Nvidia and AMD. The ultimate goal is to improve American lives through innovation and efficiency, leading to lower prices.

Trump unites tech and science for AI research

President Donald Trump launched the "Genesis Mission" on Monday, November 24, bringing together Big Tech, academics, and the government for AI research. This executive order directs the Department of Energy to create a new AI platform. It will use federal scientific data to train AI models for research in health, energy, and manufacturing. Energy Secretary Chris Wright stated the goal is to pivot private sector AI advancements to scientific discovery and lower energy prices. The mission will facilitate data sharing among DOE labs, private companies, and universities, building on existing partnerships with firms like Nvidia and Dell.

Trump launches AI Genesis Mission for US dominance

President Trump signed an executive order on Monday, November 24, creating the "Genesis Mission" to boost US leadership in AI. This initiative aims to greatly speed up scientific breakthroughs by using the Department of Energy's National Laboratories' computing power and federal data. Energy Secretary Chris Wright will establish the "American Science and Security Platform" to support research in healthcare, energy, and engineering. The mission will use AI to automate experiments and create predictive models, cutting discovery times from years to hours. Companies like Nvidia and Dell are already partnering in this effort.

Trump launches AI Genesis Mission for science

President Donald Trump signed an executive order on Monday, November 24, launching the "Genesis Mission" for artificial intelligence. This initiative aims to greatly speed up scientific discovery, boost national security, and strengthen energy independence. Michael Kratsios will lead the effort, which will expand computing resources and open up vast federal datasets. Energy Secretary Chris Wright will create the "American Science and Security Platform" to centralize AI infrastructure and train scientific foundation models. The mission encourages public-private partnerships and builds on the National AI Research Resource program.

Trump boosts AI innovation with Genesis Mission

On Monday, November 24, President Donald Trump signed an executive order to launch the "Genesis Mission," a government-wide effort to boost AI innovation. This initiative aims to transform scientific research and speed up discoveries by using federal scientific datasets to train advanced AI technologies. Trump directed the US Energy Department and National Laboratories to combine their brightest minds, powerful computers, and vast data into one cooperative research system. The mission focuses on national, economic, and health security, including biotechnology and nuclear energy, to help America win the AI race.

Zyphra trains AI model ZAYA1 using AMD MI300X

On November 24, 2025, Zyphra successfully trained its ZAYA1 AI model entirely on AMD Instinct MI300X GPUs. This large-scale Mixture-of-Experts model used AMD Pensando networking and the ROCm software stack. The MI300X's 192 GB memory allowed for simpler training and achieved ten times faster model save times. ZAYA1 performs as well as or better than other leading open models in reasoning, math, and coding. This achievement highlights the strong capabilities of AMD's hardware for complex AI workloads.

AMD hardware trains Zyphra's ZAYA1 AI model

On November 24, 2025, Zyphra, an AI infrastructure provider, successfully trained its ZAYA1 Mixture-of-Experts model using AMD Instinct MI300X GPUs. This marks the first large-scale MoE model trained solely on AMD hardware and software. ZAYA1-base showed better performance than models like Llama-3-8B and rivals Qwen3-4B. The AMD Instinct MI300X GPUs' large memory capacity simplified training and led to ten times faster model save times. This success demonstrates AMD's strong AI platform for advanced AI development.

Zyphra trains AI model using AMD Instinct GPUs

Zyphra, a leading AI company, successfully trained its first large-scale Mixture-of-Experts model using AMD Instinct GPUs. This achievement shows AMD's growing strength in the artificial intelligence sector. Zyphra's choice of AMD's Instinct platform highlights increasing trust in AMD's hardware for complex AI tasks. Training MoE models is a key step in developing more efficient and powerful AI systems. This partnership solidifies AMD's position in cutting-edge AI research and deployment.

Steven Murphy discusses AI for good

Steven Murphy, President of Ontario Tech University, discusses how artificial intelligence is rapidly changing the world. He argues that Canada should focus on developing AI for good, balancing excitement with caution. Murphy believes an ethical-by-design approach strengthens progress by addressing privacy, data management, and bias. Ontario Tech University already uses AI to help detect Type 2 diabetes and diagnose sepsis. Canada can lead globally in responsible AI by applying its values of careful planning and due diligence.

Canada can lead in ethical AI development

Steven A. Murphy, President of Ontario Tech University, emphasizes that artificial intelligence is changing the world faster than any past technology. He urges Canada to lead in developing AI responsibly, using its values of careful planning to create ethical guidelines. Ontario Tech University already uses AI to detect Type 2 diabetes via voice and help diagnose sepsis. Murphy believes an "ethics-by-design" approach strengthens progress by tackling issues like privacy and bias. Canada can find a middle ground between strict regulation and unchecked innovation, ensuring AI serves people and the planet.

Man builds AI Grandpa Bot for his children

Dr. Muhammad Ahmad from Bellevue, Washington, created "Grandpa Bot," an AI simulation of his late father. His father died in 2013, and Ahmad wanted his children to connect with the grandfather they never met. He used his father's recordings, letters, and stories to build a text-based conversational agent that mimics his dad's humor and personality. Ahmad hopes to add his father's replicated voice to the bot by the end of the year. He sees it as a modern way to remember loved ones, though he cautions against becoming too dependent on such technology.

AI and LLMs can measure mental health

On November 24, 2025, Dr. Lance B. Eliot explained that generative AI and large language models (LLMs) can effectively measure human mental health on a large scale. Unlike simple surveys, AI can conduct real-time, responsive assessments using text, video, images, and even facial expressions. This new method allows for a deeper understanding of mental health across society. Many people already use AI for mental health advice, providing a chance to analyze population trends. AI can also reach out to people for mental health checks, with their permission.

Dutch experts propose plan to lead in AI

On November 24, 2025, over fifty Dutch experts called for an ambitious "AI Delta Plan" to help the Netherlands lead in the global artificial intelligence race. They warn that without action, the country risks relying on foreign AI technology. The plan suggests appointing a State Secretary for AI, creating new AI research institutes, and designating a city for self-driving vehicle experiments. It also proposes boosting AI infrastructure with more data centers and thousands of GPUs, including a new 200 million euro AI factory in Groningen. The experts believe the Netherlands has the right ingredients but needs a focused national effort.

Google Gemini 3 versus Gemini 2.5 benchmark

This 2025 benchmark compares Google's Gemini 3 and Gemini 2.5 AI models, both with 1 trillion parameters. Gemini 3 uses "Dynamic Routing" for quick context switching and is twice as fast, scoring 90% on MMLU tests. It is currently in a paid closed preview for complex projects like enterprise app building. Gemini 2.5 uses a "Fixed Pathway," scoring 85% on MMLU, and offers a free tier for smaller tasks like code snippets or transcribing short audio clips. Tests showed Gemini 3 significantly faster, especially for larger tasks, making it better for professionals needing speed and cross-disciplinary knowledge.

Oklahoma appoints Tai Phan as first AI chief

Oklahoma's information technology division announced on Friday that Tai Phan is now the state's first chief AI and technology officer. Phan, who is also the chief technology officer, will lead Oklahoma's artificial intelligence efforts. This new role was recommended by an AI task force to help the state use AI to improve government services. Phan will set ethical standards and best practices for AI, focusing on privacy, transparency, fairness, and accountability. The task force also suggested faster access to funding to keep up with AI's rapid development.

Untangling AI capacity announcements

On November 24, 2025, Dell'oro Group analyzed the many AI capacity announcements, noting that many overlap or are overstated. The article explains that understanding actual capacity and realistic timelines is more important than just tracking new headlines. For example, OpenAI's 26 gigawatt chip-related announcements include its 10 gigawatt Stargate campus in Abilene, Texas. Much of OpenAI's capacity is likely linked to Microsoft's AI factories. Multiple companies often announce different parts of the same large project, making it hard to see the true scale of new AI infrastructure.

YOVENDO.AI launches real-time sales call intelligence

On November 24, 2025, Italian startup YOVENDO.AI unveiled a new system that brings real-time intelligence to sales calls. This technology enters calls live, detecting conversational patterns and providing immediate, actionable guidance. It measures empathy, finds timing errors, and understands objections as they happen, transforming sales conversations into measurable data. Teams using YOVENDO.AI have seen significant improvements, such as a 28 percent increase in call retention for an insurance group. The company aims to show that sales success comes from process and consistent behavior, not just natural talent.

Sources

NOTE:

This news brief was generated using AI technology (including, but not limited to, Google Gemini API, Llama, Grok, and Mistral) from aggregated news articles, with minimal to no human editing/review. It is provided for informational purposes only and may contain inaccuracies or biases. This is not financial, investment, or professional advice. If you have any questions or concerns, please verify all information with the linked original articles in the Sources section below.

AI Policy Government AI Initiatives Scientific Research Artificial Intelligence Supercomputing National Laboratories Data Sharing Public-Private Partnerships AI Ethics Responsible AI AI Applications Healthcare AI Energy AI Manufacturing AI National Security Economic Security AI Hardware GPUs AI Models Large Language Models Mixture-of-Experts AI Infrastructure Data Centers AI Innovation Predictive Models Automated Experiments Conversational AI Mental Health AI Sales Intelligence AI Self-Driving Vehicles Benchmarking AI AI Governance Data Privacy Bias in AI Genesis Mission Google Gemini OpenAI Zyphra Nvidia AMD Dell

Comments

You might also like

AI Breakthroughs and Challenges: Revolutionizing Industries and Workforce

Researchers address AI accuracy issues, innovative applications, and workforce transformation, while...

Artificial Intelligence Revolutionizes Tech Landscape with Breakthroughs and Layoffs

Discover the latest AI advancements, from deciphering ancient scrolls to OpenAI's potential hardware...

openai, nvidia and meta Updates

NVIDIA invests $100 billion in OpenAI. Meta and Google commit $560 billion to AI development by 2025...