•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
TAO, a cryptocurrency associated with the Bittensor network, is drawing increased attention from analysts who describe it as combining elements of Bitcoin’s scarcity, the S&P 500’s self-adjusting structure, and Nvidia’s AI upside in a single asset. TAO is currently trading around $250, with a market capitalization of $2.7 billion. The token runs on Bittensor, which supports 128 competing AI subnets and has a fixed supply cap of 21 million tokens.
Central to TAO’s investment thesis is its scarcity framework and the way Bittensor allocates emissions across subnets. The network is designed to emulate Bitcoin’s hard cap, with the first halving scheduled for December 2025. That event is expected to reduce daily emissions from 7,200 to 3,600 TAO.
Supply is also described as constrained by locking: about 67% of circulating supply is locked, leaving an estimated liquid float of roughly 3 million TAO.
Analysts point to Bittensor’s subnet competition as a key differentiator. The network allows 128 subnets to compete for emissions, with weaker subnets losing allocations while stronger ones attract more stake and growth. Stake-based voting is used to determine survivorship, which is framed as a market-driven process rather than a committee-driven one. The mechanism is likened to the S&P 500’s automatic index adjustment over time.
Institutional interest is rising, with multiple ETF applications currently pending before the U.S. Securities and Exchange Commission (SEC). Grayscale’s AI Fund recently allocated 43% of its TAO holdings, making TAO the fund’s largest position. Grayscale and Bitwise are among the firms with ETF applications pending with the SEC.
Other participants are also described as accumulating TAO. Corporate treasury firm xTAO is accumulating, and Yuma has staked $691 million into the network.
Adoption and usage are also highlighted through ecosystem activity. Venice.ai, co-founded by Erik Voorhees, reportedly has over one million paying users and trained its flagship model on Bittensor’s Subnet 4 using Targon Compute.
Macrocosmos, operating on Subnet 9, is described as aiming for a 70-billion-parameter distributed training run, characterized as a potential world first.
The TAO Institute launched in April 2026 and introduced a Subnet Risk Index intended for institutional allocators. Separately, Nvidia’s Jensen Huang has publicly endorsed decentralized training, a concept positioned as central to Bittensor’s approach.
Analysts also argue that higher network usage can reduce selling pressure over time. They note that stronger competition can recycle more TAO, while improved usage is expected to lower net issuance. Combined with the halving schedule, this is described as potentially creating a feedback loop that links growth to supply dynamics.
Andy ττ, an analyst cited in the coverage, said Bitcoin’s post-halving decade produced outsized returns based on scarcity and described TAO as the first liquid asset to combine Bitcoin’s scarcity, the S&P 500’s self-pruning mechanism, and Nvidia’s AI upside, trading at around $250.
Premium gym chains are entering a “golden era” that is ending or already in decline, as rising operating costs collide with shifting consumer preferences toward more flexible, community-based ways to exercise. Long-term memberships are shrinking, margins are pressured by higher rents and facility expenses, and competition from smaller, more personalized…
