By using this site, you agree to the Privacy Policy and Terms of Use.
Accept
AIModelKitAIModelKitAIModelKit
  • Home
  • News
    NewsShow More
    SpaceXAI’s Grok Tool Uploading Users’ Entire Codebase to Cloud Storage: What You Need to Know
    SpaceXAI’s Grok Tool Uploading Users’ Entire Codebase to Cloud Storage: What You Need to Know
    4 Min Read
    New York Leads the Way: First State to Enforce One-Year Moratorium on New AI Data Centers
    New York Leads the Way: First State to Enforce One-Year Moratorium on New AI Data Centers
    4 Min Read
    AI Replacing New York Nurses: Why Patients Should be Concerned About Quality of Care
    AI Replacing New York Nurses: Why Patients Should be Concerned About Quality of Care
    5 Min Read
    Navigating AI Agent Crawlers and Cloudflare’s New Rules: A Comprehensive Guide
    Navigating AI Agent Crawlers and Cloudflare’s New Rules: A Comprehensive Guide
    5 Min Read
    How Apple’s Self-Driving Car Program Paved the Way for Advanced AI Chip Technology
    How Apple’s Self-Driving Car Program Paved the Way for Advanced AI Chip Technology
    4 Min Read
  • Open-Source Models
    Open-Source ModelsShow More
    Overcoming Inference Bottlenecks: Speeding Up Complex AI Search with Retrieve-for-Train
    Overcoming Inference Bottlenecks: Speeding Up Complex AI Search with Retrieve-for-Train
    5 Min Read
    ToolGrad: Generate Efficient Tool-Use Datasets Using Textual Gradients
    ToolGrad: Generate Efficient Tool-Use Datasets Using Textual Gradients
    5 Min Read
    Enhancing Genomic Prediction in Underserved Populations through Transfer Learning
    Enhancing Genomic Prediction in Underserved Populations through Transfer Learning
    5 Min Read
    Discover TimesFM-3: A Zero-Shot Foundation Model for Enhanced Multivariate Forecasting
    Discover TimesFM-3: A Zero-Shot Foundation Model for Enhanced Multivariate Forecasting
    5 Min Read
    GlucoFM: Advanced Foundation Model for Continuous Glucose Monitoring Insights
    GlucoFM: Advanced Foundation Model for Continuous Glucose Monitoring Insights
    5 Min Read
  • Guides
    GuidesShow More
    Your Comprehensive Guide to Practical Constraint Decoding: Basics and Applications
    Your Comprehensive Guide to Practical Constraint Decoding: Basics and Applications
    6 Min Read
    KDnuggets Weekly Data Science News Roundup: Highlights from July 20, 2026
    KDnuggets Weekly Data Science News Roundup: Highlights from July 20, 2026
    4 Min Read
    Unlock Your AI Potential with Kaggle and Google’s Free 5-Day Agentic AI Course
    Unlock Your AI Potential with Kaggle and Google’s Free 5-Day Agentic AI Course
    6 Min Read
    Top 5 High-Performance MCP Servers for Optimal Agentic Development
    Top 5 High-Performance MCP Servers for Optimal Agentic Development
    6 Min Read
    Top 5 Free Resources for Understanding Agentic AI: Unlock Your Knowledge
    Top 5 Free Resources for Understanding Agentic AI: Unlock Your Knowledge
    6 Min Read
  • Tools
    ToolsShow More
    AWS Crowned Leader in The Forrester Wave: AI Infrastructure Solutions, Q4 2025 Report
    AWS Crowned Leader in The Forrester Wave: AI Infrastructure Solutions, Q4 2025 Report
    5 Min Read
    Unlock Agentic Coding: Experimenting with Qwen 3.8-Flash-Next on NVIDIA GB300 NVL72
    Unlock Agentic Coding: Experimenting with Qwen 3.8-Flash-Next on NVIDIA GB300 NVL72
    6 Min Read
    Unlock Agentic Coding: Experimenting with Qwen 3.8 Flash-Next 176B Model on NVIDIA GB300 NVL72
    Unlock Agentic Coding: Experimenting with Qwen 3.8 Flash-Next 176B Model on NVIDIA GB300 NVL72
    5 Min Read
    Optimizing LFM2.5 Q4_0 Checkpoints through Quantization-Aware Distillation Techniques
    Optimizing LFM2.5 Q4_0 Checkpoints through Quantization-Aware Distillation Techniques
    4 Min Read
    Deploy Qwen 3.8-2.4T-A95B: A Configurable 2.4T Parameter Model on NVIDIA GB300 NVL72 for Enhanced Reasoning
    Deploy Qwen 3.8-2.4T-A95B: A Configurable 2.4T Parameter Model on NVIDIA GB300 NVL72 for Enhanced Reasoning
    6 Min Read
  • Events
    EventsShow More
    Jensen Huang at Dreamforce: ‘Now We Can Know Everything and Achieve Anything’
    Jensen Huang at Dreamforce: ‘Now We Can Know Everything and Achieve Anything’
    5 Min Read
    Essential Strategies for Preparing Students for a Career in Quantum Computing
    Essential Strategies for Preparing Students for a Career in Quantum Computing
    5 Min Read
    Skild AI Leverages NVIDIA’s Physical AI to Enable Robots to Learn New Tasks from Just One Video
    Skild AI Leverages NVIDIA’s Physical AI to Enable Robots to Learn New Tasks from Just One Video
    6 Min Read
    Top 4 Mistakes New Teachers Make and Proven Strategies to Overcome Them
    Top 4 Mistakes New Teachers Make and Proven Strategies to Overcome Them
    5 Min Read
    NVIDIA Set to Acquire Hugging Face: What This Means for AI Development
    NVIDIA Set to Acquire Hugging Face: What This Means for AI Development
    5 Min Read
  • Ethics
    EthicsShow More
    Tech Leaders Demand ‘AI Slowdown’: What Would It Mean for the Future of Artificial Intelligence?
    Tech Leaders Demand ‘AI Slowdown’: What Would It Mean for the Future of Artificial Intelligence?
    7 Min Read
    The AI Industry Faces Uncertainty: What Are the Next Steps?
    The AI Industry Faces Uncertainty: What Are the Next Steps?
    4 Min Read
    Understanding Google Ad Tech Remedies: Why They Matter for Your Business
    Understanding Google Ad Tech Remedies: Why They Matter for Your Business
    4 Min Read
    How AI is Transforming China’s Disinformation Tactics in Taiwan
    How AI is Transforming China’s Disinformation Tactics in Taiwan
    6 Min Read
    Why AI Researchers Are Concerned About Machines Posing a Threat to Humanity
    Why AI Researchers Are Concerned About Machines Posing a Threat to Humanity
    6 Min Read
  • Comparisons
    ComparisonsShow More
    InternBootcamp: Enhancing LLM Reasoning Through Verifiable Task Scaling Techniques
    InternBootcamp: Enhancing LLM Reasoning Through Verifiable Task Scaling Techniques
    4 Min Read
    Enhancing Anomaly Detection in Collider Experiments through Contrastive Learning for Better Interpretability
    Enhancing Anomaly Detection in Collider Experiments through Contrastive Learning for Better Interpretability
    6 Min Read
    Exploring the Impact of Quantization on Self-Explanations in Large Language Models: Can LLMs Explain Themselves?
    Exploring the Impact of Quantization on Self-Explanations in Large Language Models: Can LLMs Explain Themselves?
    5 Min Read
    CytoNet: A Foundation Model for Understanding the Human Cerebral Cortex at Cellular Resolution
    CytoNet: A Foundation Model for Understanding the Human Cerebral Cortex at Cellular Resolution
    5 Min Read
    Optimizing Nonconvex-Nonconcave Min-Max Problems with a Limited Maximization Domain: Insights from [2110.03950]
    Optimizing Nonconvex-Nonconcave Min-Max Problems with a Limited Maximization Domain: Insights from [2110.03950]
    5 Min Read
Search
  • Privacy Policy
  • Terms of Service
  • Contact Us
  • FAQ / Help Center
  • Advertise With Us
  • Latest News
  • Model Comparisons
  • Tutorials & Guides
  • Open-Source Tools
  • Community Events
© 2025 AI Model Kit. All Rights Reserved.
Reading: Jensen Huang at Dreamforce: ‘Now We Can Know Everything and Achieve Anything’
Share
Notification Show More
Font ResizerAa
AIModelKitAIModelKit
Font ResizerAa
  • 🏠
  • 🚀
  • 📰
  • 💡
  • 📚
  • ⭐
Search
  • Home
  • News
  • Models
  • Guides
  • Tools
  • Ethics
  • Events
  • Comparisons
Follow US
  • Latest News
  • Model Comparisons
  • Tutorials & Guides
  • Open-Source Tools
  • Community Events
© 2025 AI Model Kit. All Rights Reserved.
AIModelKit > Events > Jensen Huang at Dreamforce: ‘Now We Can Know Everything and Achieve Anything’
Events

Jensen Huang at Dreamforce: ‘Now We Can Know Everything and Achieve Anything’

aimodelkit
Last updated: September 16, 2026 8:00 am
aimodelkit
Share
Jensen Huang at Dreamforce: ‘Now We Can Know Everything and Achieve Anything’
SHARE

NVIDIA and Salesforce: A Transformative Partnership at Dreamforce 2026

At Salesforce Dreamforce 2026, NVIDIA’s founder and CEO, Jensen Huang, made a powerful statement that resonated with attendees: “Know everything. Do anything.” This slogan encapsulated the spirit of innovation and progress that marked the event, particularly with the unveiling of Koa, Salesforce’s groundbreaking CRM reasoning model, built on the advanced NVIDIA Nemotron 3 Super technology.

Contents
  • A Vision for the Future
  • Elevating AI Safety Standards
  • The Symbiosis of AI and Job Market
  • Salesforce’s Strategic Implementation of Koa
  • Training and Performance Metrics
  • Immediate Applications and Future Outlook

A Vision for the Future

In his address, Huang painted a vivid picture of technological evolution, comparing the profound impacts of electricity, the internet, and now artificial intelligence (AI) on society. “With artificial intelligence as an infrastructure layer across the planet, we can know everything and do anything,” he declared, signaling a new paradigm in how businesses can leverage AI within their operations.

Elevating AI Safety Standards

During a candid exchange with Salesforce CEO Marc Benioff about AI safety, Huang was unequivocal in his stance. He emphasized that “Safety is paramount in a lot of ways. It’s job one.” This commitment to safety extends beyond mere regulatory compliance; Huang described it as an engineering responsibility that every company must undertake before launching any AI-driven product or service.

His message was clear: innovation and safety need not be mutually exclusive. “You could definitely have both at the same time,” he remarked, urging companies to prioritize rigorous testing and safety measures, ensuring their products are reliable and secure.

The Symbiosis of AI and Job Market

Addressing concerns that AI might threaten jobs, Huang pushed back strongly against the prevailing narrative. Instead, he framed AI as an enabler that enhances productivity and opens new avenues for growth. “The sky’s the limit,” he stated, encouraging businesses to actively engage with AI technology rather than fear it.

More Read

UK Prime Minister and NVIDIA CEO Ignite AI Revolution in Europe
UK Prime Minister and NVIDIA CEO Ignite AI Revolution in Europe
Understanding Struggly: A Guide for Teachers to Enhance Math Instruction
Essential AI Resources for District Leaders: Empowering Educational Leadership
Revolutionize Engineering with NVIDIA AI Physics: Boost Performance by 500x
Enhance Security Operations Readiness: Hack The Box Unveils AI-Powered Threat Range Cyber Simulation Platform

Additionally, Huang predicted a future where every enterprise would embody AI-driven capabilities, coining the term “agentic enterprise” for organizations that fully embrace AI technologies to boost efficiency and innovation.

Salesforce’s Strategic Implementation of Koa

Koa, a trailblazing model, was created using a synthetic dataset sourced from nearly three decades of enterprise CRM data. By utilizing NVIDIA’s open-source Nemotron technology, Salesforce was able to customize and run Koa entirely within its own infrastructure, ensuring that no customer data was exposed during the training or inference phases. This commitment to data privacy was emphasized by Rohan Kumar, Salesforce’s President of Platform and Engineering.

Training and Performance Metrics

The innovative facets of Koa extend beyond its infrastructure. The model was developed using advanced methodologies such as supervised fine-tuning and reinforcement learning, leveraging NVIDIA NeMo RL, NeMo Gym, and NeMo AutoModel. Training incorporated synthetic scenarios across more than 14 industries, including manufacturing, healthcare, and financial services.

In Salesforce’s proprietary CRM Bench—an evaluation suite for real-world tasks—Koa has already demonstrated exceptional capabilities, matching or surpassing leading models while achieving three times fewer errors in completing tasks like opportunity updates and case routing.

Immediate Applications and Future Outlook

Koa is not just a concept; it is already active within Salesforce’s ecosystem. As of now, it powers an employee agent in Slack and is set to enter customer pilots in October, with organizations like Formula 1 and UChicago Medicine among the first to adopt this technology. General availability is anticipated in U.S. regions by winter 2026, marking a significant step forward in AI-supported CRM functionalities.

Attendees at Dreamforce walked away with a renewed sense of excitement and clarity about the future role of AI in business. With transformative innovations like Koa, the landscape is shifting rapidly, and the integration of AI in CRM applications promises to redefine how businesses operate and engage with customers.

Inspired by: Source

Empowering Veteran Students: Effective Teaching Strategies in Technology and Learning
Understanding Prisms: How to Effectively Use Them in STEM Education
Unleashing AI Growth: How NVIDIA is Shaping the Future Everywhere, All at Once
Boosting AI Storage Performance with RDMA for S3-Compatible Solutions: A Comprehensive Guide
How NVIDIA and SAP Are Revolutionizing the Physical World with AI Agents

Sign Up For Daily Newsletter

Get AI news first! Join our newsletter for fresh updates on open-source models.

By signing up, you agree to our Terms of Use and acknowledge the data practices in our Privacy Policy. You may unsubscribe at any time.
Share This Article
Facebook Copy Link Print
Previous Article Tech Leaders Demand ‘AI Slowdown’: What Would It Mean for the Future of Artificial Intelligence? Tech Leaders Demand ‘AI Slowdown’: What Would It Mean for the Future of Artificial Intelligence?

Stay Connected

XFollow
PinterestPin
TelegramFollow
LinkedInFollow

							banner							
							banner
Explore Top AI Tools Instantly
Discover, compare, and choose the best AI tools in one place. Easy search, real-time updates, and expert-picked solutions.
Browse AI Tools

Latest News

Tech Leaders Demand ‘AI Slowdown’: What Would It Mean for the Future of Artificial Intelligence?
Tech Leaders Demand ‘AI Slowdown’: What Would It Mean for the Future of Artificial Intelligence?
Ethics
Overcoming Inference Bottlenecks: Speeding Up Complex AI Search with Retrieve-for-Train
Overcoming Inference Bottlenecks: Speeding Up Complex AI Search with Retrieve-for-Train
Open-Source Models
The AI Industry Faces Uncertainty: What Are the Next Steps?
The AI Industry Faces Uncertainty: What Are the Next Steps?
Ethics
Understanding Google Ad Tech Remedies: Why They Matter for Your Business
Understanding Google Ad Tech Remedies: Why They Matter for Your Business
Ethics
//

Leading global tech insights for 20M+ innovators

Quick Link

  • Latest News
  • Model Comparisons
  • Tutorials & Guides
  • Open-Source Tools
  • Community Events

Support

  • Privacy Policy
  • Terms of Service
  • Contact Us
  • FAQ / Help Center
  • Advertise With Us

Sign Up for Our Newsletter

Get AI news first! Join our newsletter for fresh updates on open-source models.

AIModelKitAIModelKit
Follow US
© 2025 AI Model Kit. All Rights Reserved.
Welcome Back!

Sign in to your account

Username or Email Address
Password

Lost your password?