Google is preparing to introduce a dedicated cryptocurrency security platform for banks and institutional investors, aiming to strengthen digital asset protection as cyber threats become more sophisticated. The company is expected to showcase Google Multi-Party Compute (MPC) at its first digital assets symposium in New York City on Wednesday.
The platform is currently undergoing private testing, with several major financial institutions and selected partners evaluating the technology. Google told CoinDesk that Alphabet has also begun using the system to protect its own digital assets on the Ethereum and Solana blockchains.
Google has not disclosed which financial institutions are participating in the trials or specified which assets Alphabet has secured. The company expects to launch the product commercially in early 2027 but has yet to announce its pricing.
Multi-party computation is already used across the digital asset industry to strengthen key security. The technique divides a private key into separate cryptographic shares held across different systems, allowing transactions to be authorized collectively without storing the entire key in one place. Google previously explored this approach in 2023, when it published guidance on developing MPC-based digital asset systems using its Confidential Space infrastructure.
The company said Google Multi-Party Compute will be its first dedicated MPC custody product for cryptocurrencies. Existing offerings, including Google Cloud Key Management Service (KMS) and Cloud HSM, help organizations manage cryptographic keys. The new platform brings together distributed transaction signing, policy enforcement and key recovery in a service specifically designed for digital assets.
The initiative comes as financial institutions increasingly explore blockchain applications for tokenized securities, payment processing and cryptocurrency custody. These activities require strong transaction controls because blockchain transfers generally cannot be reversed after confirmation.
“Digital asset adoption and tokenization are at an inflection point. As regulated financial institutions begin adopting blockchains and tokenization, they need scalable, proven infrastructure to secure their clients’ assets,” Rich Widmann, Google Cloud’s director of digital assets strategy, told CoinDesk.
Securing private keys remains a major concern for organizations entering the sector. If an attacker gains access to a wallet’s private key, the funds it controls could be transferred without authorization.
According to Chainalysis, private-key compromises were responsible for 43.8% of cryptocurrency stolen in 2024. DeFiLlama data showed that hacks and exploits across blockchain projects had resulted in $16.69 billion in cumulative losses by June this year, with around 40% attributed to compromised private keys.
Recent attacks highlight the consequences. In March, stablecoin protocol Resolv suffered a breach involving a compromised private key associated with a privileged account, combined with weaknesses in its minting mechanism. The attacker extracted approximately $25 million. In June, hackers compromised the private keys of a Humanity Protocol foundation member and stole more than $32 million from at least 17 wallets. The incident also sent the project’s token down by over 80%, illustrating how security breaches can affect market value as well as directly stolen funds.
Google’s new system seeks to reduce dependence on a single signing key by distributing cryptographic control among multiple participants or computers. Transactions require the relevant key shares to cooperate, rather than relying on one complete private key.
The company says the full key is never stored in one location, helping limit the consequences of an individual machine being compromised. Google also states that customers maintain control over their key material and that the company cannot access it in unencrypted form, even when key shares reside on Google’s infrastructure.
AI and quantum computing raise new security questions
Google is also framing the service as a response to evolving cyber risks, including attacks that use artificial intelligence and potential future advances in quantum computing.
AI tools are enabling some cryptocurrency fraud operations to scale, while researchers are investigating whether future technological advances could undermine the cryptographic protections used by digital asset networks.
Chainalysis found that cryptocurrency scams associated with AI-tool vendors generated an average of $3.2 million per operation in 2025, compared with $719,000 for scams without such links. The difference is approximately 4.5 times, although the findings do not prove that AI directly caused the higher proceeds.
Ethereum researcher Justin Drake has warned that advances in AI could eventually reveal mathematical shortcuts capable of breaking the digital signatures used to protect Bitcoin and Ethereum wallets. In a worst-case scenario, Drake suggested such a development might occur within months rather than years, potentially before sufficiently powerful quantum computers become available.
So far, no AI system has demonstrated the ability to break the cryptographic signatures securing Bitcoin or Ethereum wallets. AI-assisted discovery of weaknesses in poorly protected software is also different from breaking the underlying cryptographic algorithms. Google’s current MPC system focuses on threats such as stolen credentials and unauthorized transfers, rather than a hypothetical breakthrough against those algorithms.
The platform can screen proposed transactions against institutional rules before signing takes place. Organizations could establish transfer limits, restrict destination wallets or require approvals from multiple departments. These safeguards may reduce unauthorized activity, but they cannot guarantee protection when legitimate users are deceived into approving fraudulent transactions.
Quantum computing represents a separate, longer-term concern. Google researchers have previously estimated that a sufficiently advanced quantum computer could theoretically derive a Bitcoin private key from an exposed public key in about nine minutes under assumptions involving advanced preparation. No current quantum computer can carry out such an attack, but approximately 6.9 million BTC have reportedly been identified as potentially vulnerable because their public keys have been exposed.
Google says its architecture is intended to accommodate future post-quantum cryptography upgrades. These techniques are designed to protect digital signatures against attacks from sufficiently powerful quantum computers. At general availability, the service is expected to support conventional ECDSA and Ed25519 signing.
The company has not specified when post-quantum signing capabilities might be introduced. It says the system could be upgraded as blockchain networks adopt compatible cryptographic standards, without forcing institutions to rebuild their infrastructure or suspend operations.
How the platform handles transactions
A financial institution could divide signing authority among several environments, keeping one key share in Google Cloud, another on its own hardware and a third with a separate cloud provider.
When a transaction is initiated, it enters Google’s Confidential Space environment. A policy engine then checks the request against the institution’s rules before signing proceeds.
These policies can define maximum transfer amounts, approved destination addresses and requirements for authorization from multiple teams. Confidential Space uses hardware-isolated computing environments to help protect sensitive information from unauthorized access.
Google says institutions can also set different signing thresholds, such as requiring two of three participants or five of seven participants to jointly generate a valid transaction signature.
This design gives banks the option to use cloud infrastructure without placing all transaction credentials under the control of one technology provider. Key shares can be distributed among a bank’s own systems, Google Cloud and other cloud services.
Google also says the platform supports rapid signing and key recovery. However, the company has not published independent performance comparisons against existing digital asset custody products.
Google moves into a competitive custody market
The planned launch brings Google into closer competition with specialist digital asset security providers such as Fireblocks, Fordefi and BitGo, which supply technology for securing and authorizing blockchain transactions.
Google believes its relationships with financial institutions and experience in cloud security could help distinguish its offering. It is also positioning itself as an infrastructure provider for banks that want to operate their own custody businesses rather than depend entirely on specialist firms that may compete for the same customers.
Even so, adopting Google’s service will not remove the need for institutions to assess third-party risks. Banks could remain dependent on Google for portions of the software, policy controls and operational infrastructure, even when cryptographic shares are spread across multiple environments.
The system’s overall security will also depend on how customers configure it, protect the participants involved in signing and restrict access to transaction approval settings.
The key question is whether institutions will view Google’s MPC service as an appropriate alternative to existing custody platforms and how much of their critical financial infrastructure they will be willing to place within Google’s ecosystem.
