At Pwn2Own Vancouver 2022, several bugs were exploited in Microsoft, Ubuntu, and Tesla products; first day saw $800K+ awarded for exploiting 16 zero-day bugs
Jonathan Greig / The Record :
Context & Ripple Effects
The 2022 contest result extends a record of Pwn2Own demonstrations against Microsoft platforms, including an earlier Edge, Windows 10, and VMware virtual-machine escape chain. It also put Tesla and Ubuntu alongside Microsoft in the competition’s target set.
Later coverage shows the format continuing to uncover flaws in Windows, Tesla, and Ubuntu at Pwn2Own Vancouver 2024, while the 2026 Berlin event expanded successful demonstrations to AI products including Codex, Cursor, and LM Studio.
First-order effects
- Security researchers collected more than $800,000 for 16 zero-day demonstrations, while Microsoft, Ubuntu, and Tesla products were shown vulnerable to multiple previously unknown bugs.
- Microsoft and Tesla face immediate security scrutiny from the contest results, with Tesla joining conventional desktop and Linux targets in the reported demonstrations.
Second-order effects
- The recurring appearance of Windows, Ubuntu, and Tesla in Pwn2Own results pressures their security teams to treat public contest findings as an ongoing testing channel rather than isolated incidents.
- Researchers and vulnerability programs gain a visible benchmark for rewarding complex exploit work as contest payouts continue across successive Pwn2Own events.
Third-order effects
- Pwn2Own is broadening from operating systems, browsers, virtualization, and vehicles toward AI software, as shown by the 2026 exploits of Codex, Cursor, and LM Studio—making exploitability across new software layers a more public competitive measure.
- If that expansion persists, vendors across endpoint, automotive, and AI categories will increasingly compete on how quickly they can absorb adversarial testing results into product security work.
The trend: Pwn2Own is evolving into a cross-platform security proving ground, tracking the industry’s shift from endpoint software toward vehicles and AI tools.