The National Institute of Standards and Technology (NIST) on Tuesday announced the completion of the third round of the Post-Quantum Cryptography (PQC) standardization process, and we are pleased to share that a submission (SPHINCS+) with Google’s involvement was selected for standardization. Two submissions (Classic McEliece, BIKE) are being considered for the next round. 

We want to congratulate the Googlers involved in the submissions (Stefan Kölbl, Rafael Misoczki, and Christiane Peters) and thank Sophie Schmieg for moving PQC efforts forward at Google. We would also like to congratulate all the participants and thank NIST for their dedication to advancing these important issues for the entire ecosystem.

This work is incredibly important as we continue to advance quantum computing. Large-scale quantum computers will be powerful enough to break most public-key cryptosystems currently in use and compromise digital communications on the Internet and elsewhere. The goal of PQC is to develop cryptographic systems that safeguard against these potential threats, and NIST’s announcement is a critical step toward that goal. Governments in particular are in a race to secure information because foreign adversaries can harvest sensitive information now and decrypt it later.  

At Google, our work on PQC is focused on four areas: 1) driving industry contributions to standards bodies;  2) moving the ecosystem beyond theory and into practice (primarily through testing PQC algorithms); 3) taking action to ensure that Google is PQC ready; and 4) helping customers manage the transition to PQC. 

Driving industry contributions to a range of standards bodies 

In addition to our work with NIST, we continue to drive industry contributions to international standards bodies to help advance PQC standards. This includes ISO 14888-4, where Googlers are the editors for a standard on stateful hash-based signatures. More recently, we also contributed to the IETF proposal on data formats, which will define JSON and CBOR serialization formats for PQC digital signature schemes. These standards, collectively, will enable large organizations to build PQC solutions that are compatible and ease the transition globally.

Moving the ecosystem beyond theory and into practice: Testing PQC algorithms

We’ve been working with the security community for over a decade to explore options for PQC algorithms beyond theoretical implementations. We announced in 2016 an experiment in Chrome where a small fraction of connections between desktop Chrome and Google’s servers used a post-quantum key-exchange algorithm, in addition to the elliptic-curve key-exchange algorithm that would typically be used. By adding a post-quantum algorithm in a hybrid mode with the existing key-exchange, we were able to test its implementation without affecting user security. 

We took this work further in 2019 and announced a wide-scale post-quantum experiment with Cloudflare. We worked together to implement two post-quantum key exchanges, integrated them into Cloudflare’s TLS stack, and deployed the implementation on edge servers and in Chrome Canary clients. Through this work, we learned more about the performance and feasibility of deployment in TLS of two post-quantum key agreements, and have continued to integrate these learnings into our technology roadmap.  

In 2021, we tested broader deployment of post-quantum confidentiality in TLS and discovered a range of network products that were incompatible with post-quantum TLS. We were able to work with the vendor so that the issue was fixed in future firmware updates. By experimenting early, we resolved this issue for future deployments.

Taking action to ensure that Google is PQC ready

At Google, we’re well into a multi-year effort to migrate to post-quantum cryptography that is designed to address both immediate and long-term risks to protect sensitive information. We have one goal: ensure that Google is PQC ready. Internally, this effort has several key priorities, including securing asymmetric encryption, in particular encryption in transit. This means using ALTS, for which we are using a hybrid key-exchange, to secure internal traffic; and using TLS (consistent with NIST standards) for external traffic. A second priority is securing signatures in the case of hard-to-change public keys or keys with a long lifetime, in particular focusing on hardware, especially hardware deployed outside of Google’s control. 

We’re also focused on sharing the information we learn to help others address PQC challenges. For example, we recently published a paper that includes PQC transition timelines, leading strategies to protect systems against quantum attacks, and approaches for combining pre-quantum cryptography with PQC to minimize transition risks. The paper also suggests standards to start experimenting with now and provides a series of other recommendations to allow organizations to achieve a smooth and timely PQC transition.

Helping customers manage the transition to PQC

At Google Cloud, we are working with many large enterprises to ensure they are crypto-agile and to help them prepare for the PQC transition. We fully expect customers to turn to us for post-quantum cloud capabilities, and we will be ready. We are committed to supporting their PQC transition with a range of Google products, services, and infrastructure. As we make progress, we will continue to provide more PQC updates on Google core, cloud, and other services, and updates will also come from Android, Chrome and other teams. We will further support our customers with Google Cloud transformation partners like the Google Cybersecurity Action Team to help provide deep technical expertise on PQC topics. 

Additional References:

• Google Cloud Security Foundations Guide
• Google Cloud Architecture Framework 
• Google infrastructure security design overview

Google Cloud’s industry-leading controls, contractual commitments, and accountability tools have helped organizations across Europe meet stringent data protection regulatory requirements for years. This commitment to supporting the  compliance efforts of European companies has earned us the trust of businesses like retailers, manufacturers and financial services providers.

As part of our continued efforts to uphold that trust, Google Cloud was one of the first cloud providers to support and adopt the EU GDPR Cloud Code of Conduct (CoC). The CoC is a mechanism for cloud providers to demonstrate how they offer sufficient guarantees to implement appropriate technical and organizational measures as data processors under the GDPR.  

Today the Belgian Data Protection Authority, based on a positive opinion by the European Data Protection Board (EDPB), approved the CoC, a product of years of constructive collaboration between the cloud computing community, the European Commission, and European data protection authorities. We are proud to say that Google Cloud Platform and  Google Workspace already adhere to these provisions. This is the first European code approved under the GDPR; it is excellent news for the industry to have a new transparency and accountability tool that helps promote trust in the cloud. 

In addition to the CoC, Google Cloud has already been certified against internationally-recognized privacy standards such as ISO/IEC 27001, ISO/IEC 27017, ISO/IEC 27018 and ISO/IEC 27701. These certifications provide independent validation of our ongoing dedication to world-class security and privacy.

This initiative reaffirms Google Cloud’s commitment to help our customers navigate their compliance journey when using our services. To learn more about how Google Cloud can help organizations with their compliance efforts, visit our Cloud Compliance resource center.

To help secure increasingly complex and dynamic cloud environments, many security teams are turning to attack path analysis tools. These tools can enable them to better prioritize security findings and discover pathways that adversaries can exploit to access and compromise cloud assets such as virtual machines, databases, and storage buckets.

Other attack path tools rely on static, point-in-time snapshots of an organization’s cloud footprint, which often contain sensitive data about their environment, how it is configured, and where the most sensitive data resides.

At Google Cloud, we are taking a different approach. We are excited to announce today at the Google Cloud Security Summit that we are adding attack path simulation to Security Command Center, our built-in security and risk management solution for Google Cloud. This new risk management capability automatically analyzes a customer’s Google Cloud environment to pinpoint where and, importantly, how vulnerable resources may be attacked, so security teams can stay one step ahead of adversaries.

We expect attack path simulation capabilities to be available in Security Command Center Premium later this summer.

Unlike some third-party security products, Security Command Center continuously scans an organization’s cloud environment gathering near real-time data about cloud resources and security vulnerabilities. Our attack path simulation engine uses this information to automatically generate and render high-risk attack paths, without the hands-on toil of having to repeatedly run manual queries.

A different approach to attack path analysis

Other attack path analysis tools involve significant operational toil. The static, point-in-time snapshots that these tools generate have to be sent to an external provider, which can add risk. Then security teams have to follow up with complex queries before they can identify likely attack paths. 

Google Cloud’s advanced attack simulation engine leverages our first-party, agentless visibility of Google Cloud assets, the relationships between assets, and the current state of defenses. Attack path simulation is fully automated with no need to manually run queries. Simulations run in the Google Cloud environment, and do not send snapshots outside your environment, avoiding exposure of sensitive information.

See what attackers can see

Effective attack path analysis should mimic how a real-world attacker can reach and compromise high value resources. This is why Security Command Center simulates how attackers try many different ways to infiltrate a cloud environment. SCC then generates attack path graphs to give defenders insight into how adversaries could exploit a single security weakness or various combinations of security vulnerabilities to access valuable assets. SCC also provides detailed information on how to remediate issues and shore up defenses based on its findings

We are delivering combined attack path simulation and analysis as a managed service. There are no agents to install or manage. Results automatically reflect changes in your organization’s Google Cloud environment. Because our attack path simulations are conducted on models of an organization’s cloud resources, there is no performance or operational impact to the live production environment.

Better security prioritization

Security Command Center automatically computes an attack exposure score for misconfigurations and vulnerabilities that expose valuable resources to attackers. The score is a measure of cyber risk. It takes into account how exposed valued resources are, and the paths of least resistance for attackers to reach those resources. 

Security teams can use these scores to prioritize remediation efforts and improve their overall risk posture.

How attack path analysis has already reduced risk for customers 

Dozens of customers have already used our attack path capabilities in private Preview to improve their security posture and reduce their operational risk.

Security Command Center alerted one customer to a finding with a high attack exposure score. The finding was related to a service account whose keys were not being rotated. After reviewing the attack paths related to this finding, the cloud security manager discovered that even though the service account was named “test,” it provided access to storage buckets outside of the test environment. 

If an attacker had been able to steal the credentials for this test account, they could have easily accessed production data. The security manager removed administrator privileges on the account. The attack path simulation enhanced their understanding of the severity of the finding, and helped convince them to make it a high-priority fix.  

Another customer using attack path simulation needed to assess which security findings created the greatest risk. Two findings related to the same service account with high exposure scores rose to the top of the list. The attack paths revealed that the service account had access to more than 500 storage buckets. If an attacker were to gain access to this account they would be able to read, write, or delete data across any of those buckets, many of which contained sensitive business data and confidential customer information.

Using Security Command Center’s attack path results, the security team remediated the risk by limiting permissions to the storage buckets needed for that specific role.  

Next steps

Attack path simulation capabilities are planned for availability later this summer.  We expect forthcoming enhancements to use Security AI Workbench to translate complex attack graphs to human-readable explanations of attack exposure, including impacted assets and recommended mitigations.

You can learn more about Nordnet Bank’s experience using attack path simulation at our Security Summit session. To get started with Security Command Center today, please go to the Google Cloud console.

We’re excited to announce the public preview of Certificate Manager and its integration with External HTTPS Load Balancing. Certificate Manager enables you to use External HTTPS Load Balancing with as many certificates or domains as you need. You can bring your own TLS certificates and keys if you have an existing certificate lifecycle management solution you’d like to use with Google Cloud, or enjoy the convenience of our fully Managed TLS offerings.  

Extend the security and performance of the Google network to your customers

Certificate Manager brings support for multiple certificates per customer. When  coupled with our global anycast load balancing solution with automated autoscaling and failover, you now have a powerful platform for building robust SaaS and PaaS offerings. This enables custom domain support for your customers with the lowest latency and the highest level of availability. 

Alon Kochba, the head of web performance at website-building service Wix, explained how the new features lighten their workload. 

“As a SaaS product, we need to terminate SSL for millions of custom domains and certificates. GCP’s Certificate Manager and External HTTPS Load Balancing lets us do this at the edge, close to the clients, without having to rely on our own custom solution for terminating SSL,” Kochba said.

Customers who switch to External HTTPS Load Balancing can also now protect their SaaS users from denial of service attacks, OWASP Top 10 risks, and other common Web attacks by adopting Cloud Armor.

DNS authorization

This release also now enables you to provision your Google-managed certificates with DNS-based authorizations and have them ready to use before your load-balancing production environment is fully set up. This will help streamline the migration process to Google Cloud, for example. To create a DNS authorization, use the following command:

  gcloud beta certificate-manager dns-authorizations create example-authorization --domain="example.com"

This command returns the CNAME record for _acme-challenge.example.com that you must add to your DNS configuration in the DNS zone of the target domain. This CNAME record points to a special Google Cloud domain, e.g.: “534959-1a8a-40cf-90b6-b1f5f8d22517.2.authorize.certificatemanager.goog” that is used  to verify domain ownership.

When you request a certificate based on the above authorization, Cloud Certificate Manager will work with the Certificate Authority automatically to get and later renew your certificate for that domain.

Wildcard support

This DNS-based domain control authorization also allows us to bring you support for wildcard certificates. To configure the use of wildcard certificates you first must configure the DNS authorization as we’ve indicated above. Once that has been completed, you can configure the use of a wildcard certificate using the following command. Our example below is for a top-level registered domain and its wildcard subdomains.

  gcloud beta certificate-manager certificates create example-wildcard-certificate --domains="example.com,*.example.com" --dns-authorizations="example-authorization"

Monitoring for Certificate Expiration

Another new feature that will be enabled with this product  is the ability to monitor certificate expiration with Google Cloud Logging.  Cloud Logging creates a record of certificate expiration, uses the `certificatemanager.googleapis.com/Project` monitored resource, and is represented by the following message:

  message CertificatesExpiry {
  // Expiration state of the certificate.
  enum State {
    // Unspecified state, should never be reported.
    STATE_UNSPECIFIED = 0;

    // Certificate will expire soon.
    CLOSE_TO_EXPIRY = 1;
    // Certificate is expired.
    EXPIRED = 2;
  }

  // Number of reported certificates.
  int64 count;

  // Names of reported certificates. If there are too many, the list is sampled.
  repeated string certificates = 2;

  // State of reported certificates.
  State state = 3;

  // Approximate expiration time of reported certificates.
  // Multiple certificates with close expiration time are batched 
  // together in a single log, so the timestamp is not precise.
  google.protobuf.Timestamp expire_time = 4;
}

The log message is delivered every hour and contains a sample of the certificates being close to expiry or already expired.

Pricing

The best part is that there’s no additional charge to use the Certificate Manager for the first 100 certificates. To use more than 100 certificates with the management tools, we will charge on a per-certificate, per-month pricing structure. This empowers you to scale up to as many certificates as you need, and as cost-effectively as possible. The pricing will be enabled when the solution goes to General Availability.

It is our hope that these new features, combined with the programmability offered by Certificate Manager, will enable you to simplify the way you deploy HTTPS and offer a more scalable and secure service to your customers.