How will ZetaChain advance its new AI layout in the "Great Migration" to the Solana ecosystem?

On September 20, ZetaChain's token holders voted to approve a proposal that will determine its future direction: to move ZETA and the AI application Anuma to Solana, and gradually shut down its own public chain after completing the migration of related assets.

The proposal received a support rate of 99.4% with a participation rate of 58%. The migration has not yet been executed, and specific arrangements will need to be confirmed through a second proposal. However, this vote clearly indicates that ZetaChain is preparing to end its independent Layer 1 operations and concentrate resources on AI business.

A public chain that once aimed to connect other blockchains is now preparing to move its business onto another chain. Behind this change, there is a need to recalculate certain accounts.

After going all in on AI, ZetaChain no longer needs its own public chain

This transformation had already been foreshadowed before the migration proposal appeared.

On June 1 of this year, ZetaChain announced a full pivot to AI and gradually ceased its original cross-chain interoperability functions. Since then, ZetaChain's first consumer-facing AI product, Anuma, and a private memory layer have become the new core of its business.

This strategic shift not only changed the track but also altered the problems the ZetaChain team needs to solve. The migration to Solana is a further trade-off on this transformation path.

In its original business, ZetaChain served the interaction of assets and applications between different blockchains. Its underlying network was an important component of the product, as maintaining its own chain was directly related to developing cross-chain capabilities.

In the new phase of Anuma, users are more concerned about whether the model is usable, whether context can be retained during switching, and who controls the personal information remembered by AI. Therefore, whether the underlying network still needs to be operated by ZetaChain itself and its contribution to these experiences need to be reassessed.

Public chains require maintaining consensus, validator networks, and protocol upgrades, while AI products need to continuously improve model calls, memory retrieval, and user experience. As a chain based on the Cosmos SDK, ZetaChain needs to handle upstream security announcements and patches, and each update requires coordination with dozens of independent validators.

From a resource allocation perspective, ZetaChain is redefining the scope of essential options. The team believes that operating an independent L1 is no longer helpful for advancing its private AI business; the former is becoming a fixed task with limited returns that cannot be stopped.

If it migrates to other mainstream public chains, it can reduce the burden of maintaining an independent consensus network and concentrate more resources on the product. On the other hand, the application security, key management, and data flow of Anuma still need to be handled by ZetaChain itself. Changing chains can alter the division of responsibilities but cannot replace the security work at the product level.

In addition to advantages like high speed and a thriving ecosystem, Solana's appeal also lies in its existing proxy infrastructure, which can shorten the product landing path. It already has an Agent Registry for AI proxies, providing verifiable identities and reputation records; the x402 ecosystem allows network services to charge per call, enabling proxies to pay for accessing APIs, data, and content.

These capabilities can connect with ZetaChain's AI business. If a proxy is to do tasks for users, it needs to know what it is allowed to do, what information it can read, and how to pay for calling other services. ZetaChain hopes to connect its private memory and application capabilities to this infrastructure, utilizing Solana's existing identity, wallet, and payment facilities to complete the remaining steps.

This also explains why the migration plan goes beyond just transferring ZETA across chains. As long as the original L1 is retained, the dual burden of maintaining the network and operating AI products still exists. The complete migration aims to gradually concentrate technological investment, asset circulation, and developer collaboration into the same ecosystem.

Models will iterate, but memories must remain

Understanding the motivations behind ZetaChain's decision also requires answering another question: What exactly does Anuma provide that is worth the team adjusting its entire business direction? The answer lies in a personal memory that can extend across models.

As the capabilities of large models advance, reliance on AI has become an irreversible trend. For someone who uses AI extensively over a long period, there is often a need to choose different models to accomplish specific tasks based on varying requirements. This necessitates users to repeatedly emphasize expected outcomes, wording preferences, and task alignment.

If this information is stored independently of specific models, models can more easily become tools selected by task, and the work context accumulated by users can continue to be used.

The multi-model aggregation application Anuma directly addresses this issue, making it a product entry point. It integrates multiple models into a single application, allowing users to continue using existing memories and retain personal preferences and project backgrounds when switching models. According to its product design, users can manage these memories instead of reorganizing context every time they switch models.

Taking the widely used writing capability as an example, users can first use one model to organize materials and then use another model to modify the article. If both can use the same topic background and writing preferences within the authorized scope, the cost of switching models will decrease. The long-term accumulated information will not lose its value when a model is replaced.

This provides a competitive approach independent of model capability rankings. Anuma does not need to invest huge sums to train its own models to compete on model capabilities. It only needs to prove that users will find it more convenient and coherent to use different models through it.

Focusing on model aggregation, Anuma has already gained recognition from a portion of users in its early stages. Official data shows that as of September 22, Anuma has created a total of 306,900 wallet accounts and has processed approximately 1.27 million inference requests.

The upcoming test will be the user stickiness of these accounts and whether it can continuously attract new users. This is closely related to Anuma's own capabilities: the more useful the memory capabilities, the more willing users will be to use it; conversely, if the memory frequently omits key points, references outdated information, or repeatedly appears in unrelated tasks, it will directly damage the experience. Therefore, the competitiveness of this business ultimately hinges on memory quality and whether users are willing to let it participate in more daily tasks.

In addition to memory, another user-invisible capability, "privacy," is another aspect that must be understood separately. According to Anuma's technical documentation, memory adopts a local-first storage approach, with sensitive content encrypted by wallet-derived keys, and optional cloud backups uploaded as ciphertext. The wallet here plays the role of accessing personal memories, becoming an identity and key tool within the AI product.

However, memory encryption and how models handle inputs need to be understood separately. Anuma's official website states that when calling closed-source models, the current messages and related context will still be sent to the service provider, which may apply its own data retention policies; the private mode uses open-weight models and corresponding inference infrastructure.

Beyond Anuma, there is also a developer business

If these capabilities only serve Anuma, ZetaChain's growth will primarily depend on a single application. Opening up capabilities to third parties is essential to further expand the scope of this business.

For example, a company developing a travel assistant excels in destination information and itinerary planning but still needs to handle model integration, user memory, vendor failover, and calling costs. These tasks are related to its main business, but if each company does them independently, both financial and time costs will increase.

If Anuma can provide these capabilities as a service, developers can build products around their own businesses. Users may also allow different applications to utilize existing personal preferences after authorization, without having to recreate a profile in each assistant.

In fact, Anuma has already opened development access to other applications and proxies. Its currently available SDK and documentation provide development tutorials for web chat applications, mobile applications, and proxies, supporting developers in creating applications, configuring API accounts, and utilizing capabilities such as model calls, streaming responses, and tool execution.

Further open plans involve the production system that Anuma itself is using. An engineering article released on September 14 disclosed that Anuma uses Maxim AI's open-source gateway Bifrost to connect different model providers. The team is exploring opening the routing layer so that other applications and proxies can also use vendor failover, privacy model pools, and cost accounting, combined with user-authorized shared memories.

The commercial value of this plan lies in allowing third parties to not only call models but also obtain a portion of the services needed to operate around the models. Even if users are indifferent to Anuma's aggregated models, other applications may generate demand for its underlying capabilities.

The more challenging step is to allow memories to cross application boundaries while maintaining clear authorization scopes. The same personal memory should have different access ranges for different proxies. A travel assistant may need travel preferences, while a work assistant may need project backgrounds; giving all information at once is clearly not an ideal default option.

In the R&D directions disclosed by ZetaChain in September, it includes granting and revoking memory access permissions to applications and proxies, writing related records on-chain, and settling calls through x402. A longer-term vision is to allow users to organize knowledge and methods into proxies and receive rewards when called upon.

Additionally, opening up will also bring new requirements. Revoking authorization can prevent subsequent access but cannot automatically reclaim information already received by external services; third-party developers will also be concerned about fees, interface stability, and data migration arrangements. Anuma needs to manage its own applications and developer services while providing partners with sufficient reasons to rely on it long-term.

This is also a noteworthy aspect of ZetaChain's transformation: it is preparing to exit the operation of an independent public chain but still hopes to provide foundational capabilities for other developers. However, the services provided this time are more closely aligned with specific tasks in AI applications.

After migration, ZETA still needs to prove its value

For a blockchain project, the token economic system is akin to the foundation of a building, while business transformation affects the value logic of the token. After migrating to Solana, the ZETA token needs to find its place in the new business system.

According to the official announcement, the native ZETA will be converted to SPL tokens on Solana at a 1:1 ratio, with the name, total supply, and original vesting period remaining unchanged. This plan does not involve ZETA already issued on Ethereum and BNB Chain, and the staking and reward arrangements after migration are still to be clarified.

However, specific execution will wait for a second proposal. The team needs to coordinate exchange conversions first, then determine arrangements for snapshots, claims, and chain halts; the original validation and staking will continue to operate before execution.

These arrangements address the issue of how assets will be migrated. But the more long-term question is how the token will function in the new business. Or more bluntly: Why do users still need ZETA?

Solana's network fees are paid in SOL, meaning ZETA cannot directly continue the positioning of a native public chain gas token. It needs to establish demand based on application layer usage rules. The existing ZETA Access has already provided a path: users lock ZETA to obtain points that can be used for AI calls. The team hopes that as more applications connect, this access mechanism will expand beyond Anuma, allowing ZETA to serve a broader range of private AI use cases.

However, from token locking to a sustainable business, there is still a calculation that needs to be made. Locking will reduce the market's circulating tokens during the lock period but will not generate operational income, while each model inference incurs actual costs. Whether users are willing to adopt this method long-term, how points can be redeemed, and what income covers inference expenses will all affect whether this mechanism can operate sustainably.

Moreover, the increase in third-party applications also needs to be translated into demand for ZETA through specific rules. Whether developers need to hold or lock tokens for access, how long to lock, and how fees are distributed must be gradually clarified for the relationship between application growth and tokens to become clear.

The foundation of all this still lies in whether the product itself can create sustained usage demand. Anuma is attempting to expand its usage scenarios. The recently tested Nearby feature uses personal interests and preferences to discover potentially compatible people nearby, trying to extend memory from chat to real life.

Whether for social discovery or more third-party applications in the future, the team needs to prove that long-term memory can bring features that users are willing to use repeatedly. For ZETA, these features must also form clear and sustainable token use cases.

ZetaChain's choice provides a concrete case for the relationship between public chains and applications: when the product direction changes, the infrastructure that once had to be operated independently can also be handed back to external networks. Ending the operation of an independent L1 can relieve the team of some maintenance burdens and make the success or failure of this transformation more directly dependent on the product.

Shutting down its own chain is just the beginning of this trade-off. Will users continue to use it? Will developers be willing to pay for access? And can these demands find a long-term position for ZETA?

These answers will truly determine how far ZetaChain can go after shutting down its own chain.

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.