Crypto AI and Traditional AI: Distinctly Separate Tracks Traditional AI is the backbone of technology, driving transformations from medical diagnostics to autonomous driving. It focuses on algorithm optimization, model training, and enterprise applications, with funding primarily from venture capital or corporate collaborations, in a relatively mature regulatory environment with a clear development path. Crypto AI takes a different approach, combining AI with blockchain or cryptocurrencies, exploring decentralized computing, crypto trading optimization, or token incentive ecosystems. For example, ai16z supports AI Agent development through DAO, Bittensor builds a decentralized model marketplace, and Fetch.AI focuses on decentralized machine learning networks. These projects often raise funds through token sales or DeFi, carrying innovative passion while bearing the burden of market volatility and regulatory pressure. The differences are like day and night. Traditional AI is technology-driven, steadily advancing towards enterprises and research; Crypto AI navigates between decentralization dreams and speculative shadows, with token economic volatility making it appear "less orthodox". This divide laid the groundwork for NVIDIA's selection criteria. NVIDIA Inception Program: Strict Gatekeeper of AI NVIDIA's Inception Program is a beacon for AI startups, providing cutting-edge GPU resources, technical guidance, and industry connections to help enterprises soar in generative AI and data science. The entry criteria seem relaxed - just requiring company registration and a business plan, but actually favoring a combination of technological innovation and market potential. Rumors of NVIDIA limiting "cryptocurrency" projects stem from CTO Michael Kagan's 2023 statement: cryptocurrencies are "of no value to society", far less transformative than AI. This perspective seemingly cast a shadow on crypto AI, but the selection of projects like Flux (2023) and Hive Intelligence (2025) shows NVIDIA is not entirely dismissive, but selectively choosing. NVIDIA's caution is not without reason. The 2017 cryptocurrency boom led to GPU shortages, causing gamers' complaints; in 2022, SEC fined NVIDIA $5.5 million for not disclosing mining revenue. These lessons make NVIDIA especially sensitive to risks associated with crypto-related projects, preferring to focus on AI's "orthodox" track. Why Crypto AI Struggles to Gain Favor: Multiple Considerations NVIDIA's distancing from Crypto AI is not simple rejection, but based on strategic, legal, and reputation considerations: 1. Strategic Preference: NVIDIA is committed to driving AI breakthroughs in healthcare, autonomous driving, etc. Traditional AI's stable output better aligns with its vision. Crypto AI focusing on token speculation or crypto trading might deviate from this core goal. 2. Legal Risks: Token issuance may trigger securities regulations and regulatory scrutiny. NVIDIA's SEC fine experience makes it avoid such risks, with traditional AI's compliance being more attractive. 3. Reputation Concerns: The crypto industry is often plagued by negative images of fraud and high-energy mining. As a tech giant, NVIDIA needs to maintain its brand's transformative purity, cautiously approaching potentially controversial projects. 4. Resource Allocation: With limited startup acceleration resources, NVIDIA prioritizes AI projects deeply synergistic with its GPU ecosystem, like generative AI, rather than crypto projects dependent on token economics. Nevertheless, Hive Intelligence's selection demonstrates that if Crypto AI focuses on technology with blockchain as an auxiliary, there's still a possibility of breaking through.
Bittensor should optimize the inflation mechanism, accelerate dTAO implementation to enhance subnet attractiveness, and develop a more user-friendly interface to lower technical barriers.
Fetch.AI (FET): Ecosystem Integration and Competitive Pressure
Project Overview
Fetch.AI (FET) supports DeFi, transportation, and other scenarios through a decentralized machine learning network. In 2024, it merged with SingularityNET and Ocean Protocol to form the ASI Alliance, with the $ASI token doubling in growth. FET is characterized by an open, permissionless network that attracts developers and enterprise users.
Limitations Analysis
- Merger Integration Challenges:
- The ASI Alliance aims to integrate resources to accelerate decentralized AGI, but token migration and ecosystem fusion increase complexity. Some users worry that the merger might weaken FET's independence, and uneven resource allocation in the short term could affect development progress.
- Competitive Pressure:
- In 2024, over 600 crypto AI projects emerged with a market cap exceeding $70 billion. Fetch.AI faces competition from opponents like Bittensor and NEAR Protocol. AI Agent projects in the Solana ecosystem threaten its market share with low costs and high throughput, requiring FET to further optimize performance and costs.
- Insufficient Application Deployment:
- Despite FET's applications in DeFi and transportation, its actual adoption rate is lower than expected. Compared to mature enterprise cases of traditional AI, FET's decentralized network still needs breakthroughs in commercialization and requires more flagship use cases to prove its value.
Improvement Directions
Fetch.AI needs to accelerate ASI Alliance integration, optimize network performance to address competition, and promote landmark application deployment to enhance market recognition.
Common Bottlenecks of Crypto AI and NVIDIA's Perspective
The limitations of ai16z, Bittensor, and Fetch.AI reveal common issues in crypto AI:
- Token Speculation: Inflation or opaque distribution (like TAO) weakens trust and obscures technical value.
- Technical Barriers: Blockchain integration increases development difficulty and limits non-crypto user participation.
- Market Fragmentation: Emerging new protocols require existing projects to continuously innovate to maintain competitiveness.
- Regulatory Risks: Token economics may trigger securities law review, increasing uncertainty.
These issues are directly related to NVIDIA's screening logic. The acceleration program favors technically driven, low-risk projects, while the speculative nature and complexity of crypto AI might be seen as "not of high caliber". However, ai16z's community vitality, Bittensor's subnet innovation, and FET's open network also demonstrate potential, which could be breakthrough if optimized.
Knocking on NVIDIA's Door: Crypto AI's Breakthrough Path
To win NVIDIA's favor, crypto AI needs targeted improvements:
- Focus on AI Core: Emphasize algorithmic innovation or GPU-intensive applications, downplaying token speculation.
- Lower Barriers: Develop user-friendly tools to attract traditional AI developers and expand user base.
- Transparent Governance: Optimize token distribution and community decision-making to enhance trust.
- Application Deployment: Create landmark use cases to prove the commercial value of decentralized AI.
Hive Intelligence's success demonstrates NVIDIA's openness to technological innovation. Crypto AI needs to raise the AI banner with blockchain as an auxiliary, shedding the "speculation" label.
Conclusion: From Being Shut Out to Breaking Through
NVIDIA's startup acceleration program is a trial ground for AI and crypto intersection, and its caution towards crypto AI reflects the contest between innovation and risk. The stability of traditional AI contrasts sharply with crypto AI's disruptive nature. The limitations of ai16z, Bittensor, and Fetch.AI—speculation, complexity, competitive pressure—explain why they struggle to reach "high ground". However, crypto AI's potential cannot be ignored. The dream of decentralization and the fusion of technology and economics are driving the industry forward. NVIDIA's "shutting out" is not an endpoint, but a challenge. If crypto AI can wield technology as a sword and trust as a shield, it may knock on the door and write a new chapter of AI and blockchain dancing together.
