SoftBank’s Purchase of Ampere, Future of Ed-Tech | Bloomberg Technology

Key Concepts:

• Quantum Computing
• AI (Artificial Intelligence)
• NVIDIA GTC (GPU Technology Conference)
• AI Infrastructure
• Ampere Computing Acquisition
• DeepSeek AI Model
• AI Supply Chain Dominance
• Tokenization
• High-Temperature Superconductors (HTS)
• For-Profit Education
• Pell Grants

1. NVIDIA and the AI Infrastructure:

• NVIDIA's CEO stated that the company will procure half a trillion dollars worth of electronics in the next four to five years. This is not capital expenditure but rather the cost of building and upgrading AI data centers.
• NVIDIA holds a 70% market share in AI accelerators (high-performance GPUs).
• Building AI data centers involves significant costs, including manufacturing chips (TSMC), server assembly, and packaging.
• NVIDIA is positioning itself as an "AI factory" company, providing the infrastructure for AI development.
• NVIDIA sells existing technology to the quantum computing industry to improve their machines but does not make quantum computers themselves.
• Jensen Huang believes quantum computers are a decade away from being useful.

2. Market Analysis and Investment Strategies:

• The market can be emotional, with investors selling high-performing stocks (like the "MAG Seven") during periods of panic or fear (e.g., tariff concerns).
• There is a lot of liquidity on the sidelines, and consumer sentiment is uneasy.
• Regulatory clarity, tax cuts, and resolution of tariff issues could lead to a market rally.
• Even without these factors, the economy is still growing, and corporate America is still investing in capital expenditures (CAPEX).
• Quantum computing and AI are worth taking risks for long-term returns.
• Broadcom is identified as another potential leader in AI, with significant AI revenue growth.

3. SoftBank's Acquisition of Ampere Computing:

• SoftBank is acquiring Ampere Computing for $6.5 billion, highlighting the increasing demand for compute and infrastructure.
• Ampere Computing is a leading supplier of high-performance, power-efficient processors for data centers.
• The acquisition is synergistic, allowing Ampere to continue its roadmap, including AI acceleration, as part of the SoftBank family.
• Ampere will continue its existing product line and work with existing customers.
• Ampere recognized early on the need for high performance with lower power consumption.
• Power is seen as the biggest limit to long-term growth, especially with the massive growth in AI compute, which requires 10x the power.
• The future of computing involves integrated computing in everything, with AI transparently integrated into appliances, cars, and other devices.

4. DeepSeek and the AI Model Landscape:

• DeepSeek's AI model has gained attention for performing better than its counterparts.
• Companies in China are racing to integrate the DeepSeek model.
• 01.AI is building a user interface and application layer for DeepSeek to make it useful for finance, customer service, and other corporate applications.
• The goal is to make DeepSeek knowledgeable about specific company data (e.g., CRM databases, employee records).

5. AI Supply Chain Dominance:

• A few companies dominate the AI supply chain: NVIDIA (GPUs), TSMC (chip manufacturing), makers of high-bandwidth memory, and makers of lithography machines.
• These companies wield tremendous power over the availability and deployment of AI technologies.
• Monopolies tend to last for a long time, fading due to government intervention.

6. Compute Demand and Tokenization:

• Compute demand is 100x greater than it was a year ago.
• This is measured by counting the tokens in a tokenization context (three bytes of data per token).
• DeepSeek bought a huge amount of GPUs because inference is so compute-intensive.
• Reasoning models require more tokens but can provide better outputs with fewer questions.

7. NVIDIA's Strategy and Enterprise Adoption:

• NVIDIA is laser-focused on making the most efficient token.
• Enterprises need predictability and partners to provide roadmaps and deliver solutions.
• Investments in AI are huge, and enterprises need trust in channel partners.
• NVIDIA is working with recruiting vendors on cooling plate designs and other aspects of AI infrastructure.
• Having a platform and models allows people with different skill levels to work with AI.

8. Quantum Computing and NVIDIA's Role:

• NVIDIA does not make quantum computers but provides AI supercomputer technology for error correction.
• CPUs are a big part of quantum computing.
• Quantum computing is where ARM was in 2011, potentially needing another five years to become highly reliable.

9. For-Profit Education and Campus:

• Campus is a for-profit education company that aims to provide a more affordable path to a college degree.
• It is building a new kind of two-year college where students learn from professors from top schools and transfer to four-year schools with no debt.
• The traditional community college has an average completion rate of 27%.
• AI is changing everything, and students need the right skills and frameworks.
• The focus is on adding courses that teach skills needed in an economy where AI is involved.

10. Federal Government's Role in Education:

• There are concerns about the federal government's involvement in education and potential cuts to funding.
• Pell Grants are used by the vast majority of Campus students to cover tuition.
• There is a push for more accountability and efficiency in how taxpayer dollars are spent on education.

11. Fusion Energy and Superconducting Magnets:

• Fusion energy is the next breakthrough.
• Primer is exploring the use of superconducting magnets in fusion reactors.
• High-temperature superconductors (HTS) allow for smaller and cheaper fusion devices.

Synthesis/Conclusion:

The broadcast covers a range of technology and finance topics, from NVIDIA's dominance in the AI infrastructure market and SoftBank's strategic acquisition of Ampere Computing, to the rise of DeepSeek's AI model and the challenges of AI supply chain consolidation. It also explores the potential of quantum computing, the changing landscape of for-profit education, and advancements in fusion energy technology. The key takeaway is that AI is driving significant growth and innovation across various sectors, but also presents challenges related to infrastructure, supply chain dominance, and the need for skilled talent.