Teaching AI to make tough calls—without exposing your values

Kari Jaaskelainen

12 Nov 2025 — 1 min read

Self-driving cars may face no-win crash scenarios. To build fairer AI, researchers often ask the public which option they prefer, then aggregate the answers. But your moral preferences are deeply personal—and sharing them can risk your privacy.

This study introduces a first privacy-preserving, crowd-guided way to train AI for ethical dilemmas using differential privacy (a proven method to hide individual data in statistics).

How it works

Centralized protection (VLCP, RLCP): A trusted aggregator adds carefully tuned noise to the group average, shielding either each voter or each record.
Distributed protection (VLDP, RLDP): Each person privately perturbs their own response with a personalized privacy setting before sending it.

Across synthetic and real datasets, the approach kept decision accuracy high while protecting individuals’ moral choices.

Bottom line: We can guide AI with our values—without revealing who chose what.

Paper: http://arxiv.org/abs/1906.01562v2

Register: https://www.AiFeta.com

#AI #Ethics #Privacy #DifferentialPrivacy #AutonomousVehicles #Crowdsourcing #DataProtection #ResponsibleAI

Automating GDPR Compliance: A Roadmap for Companies and Law Firms

GDPR compliance is more than checkboxes. A new roadmap from the Privatech project shows how automation and machine learning can help companies and law firms assess—and even generate—privacy compliance. * Shift the focus to data processors’ real workflows: drafting policies, mapping data uses, documenting decisions. * Break compliance into machine-ready

FPGAs for Faster, Leaner Deep Learning: A Review of CNN Accelerators

Deep learning drives image search, robots, and medical scans. Most systems lean on CPUs and GPUs. This review asks: what if we run convolutional neural networks (CNNs) on FPGAs—reconfigurable chips you can tailor to the model? * Why FPGAs: custom dataflows, low latency, and strong energy efficiency—great for cameras,

Dynamic-K: Recommendations That Know When to Stop

Most apps show a fixed number of “top” items—say 10 movies or 20 products—assuming there are always enough good options. But that’s not always true: sometimes there are few relevant items, or some users are extra picky. The result? Filler recommendations. Dynamic-K flips the script. Instead of

Teaching chatbots to stop contradicting themselves (DECODE)

Teaching chatbots to stop contradicting themselves Ever had a bot say one thing, then the opposite a few turns later? This study introduces DECODE—a new task and dataset for spotting contradictions in everyday conversations, drawn from both human-human and human-bot chats. * New data beats existing natural language inference (NLI)

How it works

Read more

Automating GDPR Compliance: A Roadmap for Companies and Law Firms

FPGAs for Faster, Leaner Deep Learning: A Review of CNN Accelerators

Dynamic-K: Recommendations That Know When to Stop

Teaching chatbots to stop contradicting themselves (DECODE)