Pwn2Own Automotive 2025 produced 49 unique zero-day vulnerabilities and paid out $886,250 across EV chargers, in-vehicle infotainment systems and other automotive devices. The result shows that researchers find real bugs quickly when they are paid to look. It does not tell you how many bugs are in any single product.
What was on the table
Trend Micro's Zero Day Initiative ran the contest. According to its final results post, the targets included EV chargers from ChargePoint, Autel, Ubiquiti, WOLFBOX and Tesla, plus in-vehicle infotainment units from Sony, Alpine and Kenwood.
A contest like this works through coordinated disclosure. Vendors get the details of a bug found in their product, and the contest sets a clock. That is a healthy process, and it gives a rare public view into how exposed new device classes are.
How to read the numbers
Forty-nine zero-days is a headline, so handle it with care.
- It is a count of unique bugs across many products, not a per-product defect rate.
- Contest entries are chosen by researchers for payout and tractability, so the sample is not random.
- A fixed bug is a good outcome. It means the process worked for that product.
- Contest rules can reduce an award when an entry relies on a previously known issue. That is a reminder that old, unpatched bugs still matter.
The honest reading is narrow: these device types had exploitable flaws that skilled researchers could find within a contest window.
Why chargers and infotainment units
Both device classes sit at an interesting boundary. A charger connects the grid, a car and a network. An infotainment unit connects a screen, a phone, a cellular link and often internal vehicle networks. Both run substantial software stacks, and both are built by supply chains that include third-party code.
That combination is why they get attention. The more code a connected device carries, the more places a flaw can live.
What device makers can take from it
- Plan for external research. A published vulnerability disclosure policy and a named contact make the difference between a clean report and a public surprise.
- Ship a real update path. A bug that cannot be patched in the field becomes a permanent exposure.
- Know your components. When a flaw lands in a library used by several products, you need to find every affected build fast.
- Test the web and network interfaces that every charger exposes, since they are reachable by design.
What operators and buyers can do
If you run chargers or fleets, ask suppliers how they handle disclosures and how long they support a product with security updates. Keep chargers on segmented networks, and keep management interfaces off the open internet.
Safeguard helps with the software inventory side: SBOM and dependency scanning for the code your team builds, and a free scan at https://gold.safeguard.sh/scan. It does not test charger hardware.
Quick check
Does 49 zero-days mean chargers are unsafe to use?
No. It means researchers found 49 unique bugs across many products during a contest. Vendors receive the details and can fix them, so read it as evidence of exposure and of a working disclosure process.
Sources
Last reviewed: 19 October 2026.
