Hong Kong Regulators Warn LinkedIn Users About AI Data Training: Key Steps to Opt Out by 2025
- Why Are Hong Kong Regulators Sounding the Alarm?
- How to Opt Out of LinkedIn’s AI Training
- The Bigger Picture: AI’s Insatiable Data Hunger
- What’s Next for Data Privacy and AI?
- FAQs: LinkedIn’s AI Training and Your Data
starting November 3, 2025, their personal data will be used to train generative AI models unless they manually opt out. The Office of the Privacy Commissioner for Personal Data (PCPD) emphasized that users must review LinkedIn’s updated privacy policy and take action to protect their information. This MOVE follows months of negotiations between regulators and LinkedIn, which initially paused its AI training plans for Hong Kong users in late 2024. Here’s what you need to know to safeguard your data—and why this debate reflects a global scramble for AI training materials as tech giants face a looming data drought.
Why Are Hong Kong Regulators Sounding the Alarm?
In October 2024, LinkedIn announced plans to leverage user profiles, resumes, public posts, and activity to train its AI systems—a policy set to roll out globally by September 2025. However, Hong Kong’s PCPD intervened, citing concerns over default opt-in settings that automatically enrolled users. After a six-month standoff (October 2024–April 2025), LinkedIn agreed to let Hong Kong users retain control. Privacy Commissioner ADA Chung Lai-ling stressed, “Users must understand how their data will be used and consciously consent—or decline.”
How to Opt Out of LinkedIn’s AI Training
LinkedIn claims only public data (not private messages) will be used, excluding minors under 18. To opt out:
- Navigate to Account Settings > Data Privacy
- Select “Generative AI Data Enhancement”
- Toggle off “Use my data for AI content creation models”
“This isn’t just a checkbox—it’s your digital autonomy,” remarked a BTCC market analyst. “Platforms like LinkedIn and Meta (which revived similar plans post-regulatory review) are banking on user inertia.”
The Bigger Picture: AI’s Insatiable Data Hunger
LinkedIn’s parent company Microsoft—a major OpenAI investor—plans to share user data across subsidiaries for AI development. But there’s a catch: industry leaders like Goldman Sachs’ data chief Neema Raphael warn that models like ChatGPT and Google’s Gemini are running out of training data. OpenAI cofounder Ilya Sutskever predicted in 2024 that AI progress WOULD “hit a wall” without new sources. “We’re scraping the barrel,” admitted a TradingView insider. “Agent-based AI—systems that act autonomously—might be the next frontier.”
What’s Next for Data Privacy and AI?
Hong Kong’s PCPD vows ongoing monitoring, aligning with global trends. Meanwhile, users face a dilemma: contribute data to fuel AI advancements or lock down their digital footprints. As one Reddit user quipped, “Your resume could end up teaching a robot to write better resumes—meta, right?”
FAQs: LinkedIn’s AI Training and Your Data
When does LinkedIn’s AI data training start?
LinkedIn will begin using eligible user data for AI training on November 3, 2025.
Which regions are affected?
Users in the UK, EU, EEA, Switzerland, Canada, and Hong Kong are included—but Hong Kong residents can opt out.
What data is used?
Public profiles, posts, resumes, and activity—not private messages or data from users under 18.
Why is Hong Kong special?
Regulators forced LinkedIn to pause plans in 2024 and mandate explicit user consent after backlash over default opt-ins.
