Opinion
30 octobre 2026
Florida DOE data breach potentially exposes SSNs of 4.9 million people
The Florida Department of Education (FLDOE) has disclosed a cybersecurity breach potentially affecting (…)
Voir →The global average cost of a data breach was $4.44M in 2025, with the average cost per compromised record at $160, as reported by IBM. Of course, the cost of cybersecurity attacks extends far beyond direct economic loss. It affects an enterprise’s reputation and trustworthiness, as well as individual clients, patients, or citizens whose personal data may be compromised. For this reason, many organizations are turning to homomorphic encryption—a relatively new type of encryption that adds noise to ciphertexts, making it more secure than its traditional counterpart. Unlike traditional encryption, which protects data at rest and in transit but requires decryption for processing, homomorphic encryption enables computations to be performed directly on encrypted data without exposing the underlying plaintext. The technology is reshaping how enterprises approach privacy, analytics, and cross-jurisdictional collaboration.
What Is Homomorphic Encryption?
The most common analogy for homomorphic encryption is a locked box containing sensitive documents. Say you wish a third party to analyze or undertake another action with your documents, but you don’t want to give them the key to open them. Traditional encryption essentially forced you to hand them the key, let them take the action you wanted, then lock the box again to keep the contents safe. Homomorphic encryption offers a new capability: it allows you to grant third-party access to the contents of the box without unlocking it. When they are done, the documents are still encrypted, but the results of their actions or calculations are reflected in the encrypted form. When you decrypt it, you see the processed result, but the third party never saw the original data.
How Homomorphic Encryption Can Be a Powerful Ally for Enterprises
Homomorphic encryption can help address key cybersecurity risks faced by enterprises from a wide array of industries. These include companies that rely on public cloud infrastructure and SaaS platforms for analytics, AI, and general business operations. It can also provide organizations with greater security when using AI and advanced analytics. Secure collaboration across departments and organizations is another key concern. Modern enterprises typically work with numerous suppliers, research groups, and sometimes, even competitors. Examples of enterprises that collaborate with various entities include healthcare providers, manufacturers, retailers, and energy companies. Homomorphic encryption is also a boon for companies concerned about CMMC compliance, including defense contractors, who must comply with strict cybersecurity regulations imposed by the U.S. Department of Defense. It also ensures that enterprises comply with privacy laws, such as the GDPR.
A Real-Life Example
To understand homomorphic encryption, it helps to see how it works in practice. Say your organization is a bank that wishes to run advanced risk and fraud analytics on customer transaction data. The tools you need to carry out these actions are hosted by a third-party cloud analytics vendor. You do not wish to share the data with them because it contains your clients’ personal information (including identities, account balances, and transaction histories). Under the traditional system, you would encrypt data and send it to the cloud. To conduct analytics, the cloud provider would need to decrypt the data in memory. During the processing stage, the data would be visible to their systems and therefore vulnerable to cyberattacks. After the analysis, the results would be re-encrypted and returned. Under homomorphic encryption, you would encrypt the data, send it to the cloud provider, and the provider would conduct analysis, fraud detection, and risk scoring directly on the encrypted data. They would then send you encrypted results—for instance, of risk metrics or fraud scores. Only you would be able to decrypt the output.
Types of Homomorphic Encryption
There are three types of homomorphic encryption. The first, partially homomorphic encryption (PHE), supports only one operation—for instance, addition or multiplication. The second, somewhat homomorphic encryption (SHE), supports a limited number of operations. The third, fully homomorphic encryption (FHE), supports arbitrary computations on encrypted data and enables the most advanced use cases.
Challenges Ahead
Homomorphic encryption offers numerous advantages, including enhanced security, improved compliance, and likely quantum-proofing. However, it has drawbacks, including limited scalability. Data volume is increasing rapidly, and homomorphic encryption incurs longer processing times, which can lead to scalability issues. This type of encryption also requires more computational resources, which can increase costs. Lastly, it takes approximately 360 times longer to encrypt data than with classic techniques, which may frustrate enterprises with large amounts of information.
Homomorphic encryption is a buzzword in cybersecurity, as it adds significant robustness to encrypted data. Allowing third parties to process data in the cloud without decrypting it ensures that private data remains private. On the downside, it can lead to longer waits and higher costs, so each enterprise must make its decisions based on the amount of data it has, its security needs, and its third-party relationships.
