Last updated on September 18th, 2026 at 04:50 pm
All companies want smarter AI. Hospitals want models that can diagnose illness earlier. Banks want fraud detection that identifies suspicious patterns across institutions. Retailers want recommendations that truly seem personalized.
But there’s a wall nobody wants to discuss: data cannot leave the building.
Rules, contracts, and competitive issues mean the most you can do is exchange data between companies. How do you train a model on data you can’t even access?
Precisely that gap, which federated learning and multi-party computation are comfortably filling and disrupting in subtle ways that, allow me to point out, are transforming the application of AI at scale.
Table of Contents
What Federated Learning Actually Does (Without the Textbook Definition)
Here’s another way of putting it: Suppose a few dozen hospitals all have records on patients and you’re trying to learn from all of their data. How do you handle that? Well, traditionally, you’d combine all the data into one megastorage system. Not gonna work. This creates a ton of data compliance problems and increases breach risk.
That’s where federated learning flips the script. It simply takes the model to the data, rather than the data to the model.
A local copy is trained on data at each hospital. Only model updates, i.e., gradients- what the model has learned get sent back to the base. The data record itself never leaves.
I’ve read this concept discussed dozens of times, but only in the abstract. Still, for the first time I felt its true significance when I saw a team of healthcare AI researchers describe how they can effectively train models on data distributed across six hospital networks without a single patient’s record crossing a physical institutional boundary. That‘s no small feat.
But as you update the model, you might reveal some information about the training data (if you’re not careful). That‘s where the second piece comes in.
Multi-Party Computation: The Part Most Articles Skip
Multi-party computation (MPC) predates federated learning, and frankly, it doesn’t get the recognition it deserves.
Main concept: Several people computing a function using their combined inputs without any one of them knowing the other’s input. The classic example is called Yao’s Millionaires’ Problem, where two people want to find out who’s richer but don’t want to tell the other their total net worth. MPC can solve this mathematically.
Signaling for AI: individual institutions contribute to a collective calculation of model updates, aggregate them, or verify the output without revealing individual values. Combined with federated learning, it seals the breach opened by sharing raw gradients.
How These Two Technologies Work Together
Federated learning solves the training distribution problem – the data stays local. MPC tackles the aggregation trust problem – even the aggregation step should not reveal too much.
In practice, this means:
- Models are encrypted or secret-shared by each user after they complete updates.
- The aggregation server combines them without decrypting each contribution.
- It’s not just the final aggregate number that’s reported back; it’s not every detail.
This is most useful when no one should see the whole picture (e.g., cross-bank fraud detection, or retail MI without revealing sales data).
Federated Learning and Multi-Party Computation in the Real World
This is no longer research territory.
Medical Applications: Google Health and DeepMind have tried federated methods for medical images. They want to train diagnostic models across hospital systems in different countries, each with distinct data protection regulations.
Mobile phones: Google’s Gboard keyboard uses federated learning to improve next-word predictions without sending any typing data off the phone. I used Gboard for several years and saw autocorrect consistently improve while making my own contribution with every keystroke I didn’t even send.
Finance: banks leverage federated and MPC-based systems in developing joint fraud schemes. Banks share pattern detection, but they don’t share customer transaction histories with other Participants.
Cross-border AI: Data residency requirements in the EU and the US differ. Federated learning can let them train the same models without moving data across jurisdictions.
Where Privacy-Enhancing Technologies (PETs) Fit Into This
Federated learning may not work alone. Along with many other tools, they fall under an umbrella called PETs (Privacy-Enhancing Technologies), which aim to minimize data while enabling computation.
PETs include:
- Differential privacy: perturbed output with a tuned ratio, so individual records cannot be reverse-calculated.
- Secure enclaves: a hardware-level protected, separated domain for executing sensitive information.
- Homomorphic encryption: how we compute directly on the encrypted data.
- MPC: computation involving several parties without revealing inputs to each other.
- Distributed model training: federated learning
The structure of federated learning and MPC is ‘privacy-preserving ’; privacy is part of how the computation is set up to happen. It is not bolted on as an afterthought.
In my research, I observed that many organizations view differential privacy as a secondary aspect of federated systems, which makes sense because the two capabilities build on each other’s safeguards.
The benefits of Homomorphic Encryption and why it is both promising and painful
Homomorphic encryption refers to a method that lets you perform computation on encrypted data, and the results, once decrypted, match what you would get if you performed the computations on unencrypted data.
This sounds like a silver bullet for collaborative AI. And in principle, it is.
In practice? It’s expensive. Machine learning on fully homomorphically encrypted data can be an order of magnitude more computationally expensive than plaintext data. For simpler aggregations/operations (like Partial or somewhat homomorphic schemes), the overhead is easier to bear, and you see it used in specific portions of FL pipelines, rather than entirely.
However, the technology is moving rapidly. We have seen a dramatic reduction in the gap over the last 3 years through hardware acceleration and algorithmic developments. IBM and Zama continue to push that hardware boundary.
My Take on Where Homomorphic Encryption Realistically Fits Today
Right now, it’s most practical for:
- Secure aggregation in federated systems (not in full training)
- Encrypted inference where a client encrypts an input and the server returns an encrypted prediction.
- Certain financial calculations where precision takes precedence over speed
Would this be suitable for training large models? Still a few years away from being feasible without hefty hardware investment.
PETs for Regulatory Compliance – Where Things Get Commercially Real
If you want to build AI products that involve your user data – wherever in the EU, UK, or anywhere else that has serious data protection regimes in place – PETs for Regulatory Compliance is no longer optional.
GDPR starts by requiring data minimization. As a result, HIPAA restricts what can exit its covered systems. Financial regulators in various jurisdictions have restricted cross-border data flows. The UK Information Commissioner’s Office has provided guidance on how PETs can meet compliance requirements.
2. Data minimization is also met directly with federated learning, where just model updates are shared and not records. MPC also helps meet access control and need-to-know requirements. They can therefore give a compliance team something tangible to reference.
What I find interesting in practice is that once legal teams understand federated architectures, they tend to love them, because the data governance story is much cleaner. Instead of negotiating complicated data-sharing agreements, you’re negotiating model-sharing agreements, which are far less regulatory-burdensome.
This is one of the less-publicized commercial benefits: federated + MPC systems can help speed up deals that would otherwise be held up at the Legal Review stage.
What Most People Get Wrong About These Technologies
A few misconceptions worth clearing up:
“Federated learning means your data is private.” Not necessarily. Without safeguards like secure aggregation, members’ inference attacks can reverse-engineer training data information from gradient updates. Differential privacy or above is required.
“MPC is just for cryptographers.” The math is complex, and the tooling has matured. Libraries such as PySyft, OpenMined, and FATE make federated + MPC configurations available to ML engineers without a cryptography PhD.
This is relevant just for giant companies. Medium-sized companies with sensitive B2B data – especially legal tech, HR analytics, and supply chain – have exceptionally tangible use cases: any situation where two organizations want to leverage learnings from others’ data without sharing sensitive information.
“Way too slow to be practical.” For certain applications, yes. For others – particularly asynchronous training, inference-only applications, and aggregation applications – the overhead on today’s hardware is acceptable.
Who Should Actually Be Paying Attention to This
If you’re a developer working on data products that involve regulated industries, this should be on your mind.
If you’re a product manager or technical founder, the pitch is straightforward: Federated learning and MPC give you something no one else can copy, a reasonable way to collaborate in the presence of sensitive data. A compelling value proposition for a healthcare, financial, or legal SaaS company, or any B2B column.
If you are technically curious, this architecture is incredibly cool. It’s a totally different paradigm that you can train a model on siloed, distributed data without having to gather the data itself.
Conclusion: The Honest Take
Federated learning and MPC aren’t without their issues, either. They do complicate the problem, and they have to be implemented correctly, and in some use cases there still may be performance constraints, for example, where end-to-end homomorphic encryption is sought.
But they point to something truly valuable: a path to collaborative AI that doesn’t put your compliance stance at risk over a data-sharing deal.
The technology stack of PETs, lean federated architecture, and selective application of MPC already shows results in show-floor systems across healthcare, finance, and mobility. The tooling is maturing. The regulatory pull to go privacy-by-design is only heading in one direction.
In the 18–35 tech community building or working around AI systems, this is good to understand now, not later. Those who learn to coordinate over data without sharing it will have a structural advantage over teams that can’t get there yet.
I’m a technology writer passionate about AI and digital marketing. I create engaging and useful content that bridges the gap between complex technology concepts and digital technologies. My writing makes the process easy and engaging. I encourage participation I continue to research innovation and technology. Let’s connect and talk technology!
