Pwn2Own Automotive 2026 Ends With 76 Zero Days and $1.047M Paid Out, A Record Signal for EV Charger and IVI Risk

What Happened at Pwn2Own Automotive 2026

Pwn2Own Automotive is designed to pressure test real world automotive technology under conditions that approximate how attackers think: constrained time, high incentives, and a focus on outcomes such as code execution, root access, or meaningful device manipulation.

In 2026, that pressure produced the largest vulnerability count the Automotive edition has recorded so far. The contest ran January 21-23 in Tokyo during the Automotive World conference, and the targets were not lab curiosities.

Scope included:

• Fully patched IVI systems
• EV charging infrastructure
• Automotive operating systems (Automotive Grade Linux)

Day-by-Day Progression

The day-by-day progression matters because it shows how quickly the field moved from "interesting exploits" to "systemic pattern."

Day	Awards	Zero Days	Cumulative
Day 1	$516,500	37	37
Day 2	$439,250	29	66
Day 3	$91,250	10	76
Total	$1,047,000	76	—

Day 1 alone awarded $516,500 for 37 unique zero days, an intense pace that would have been a full event total in earlier years. Day 2 pushed the running total to 66, surpassing the prior three-day benchmark of 49 vulnerabilities seen in previous Automotive editions.

Day 3 also highlighted collisions - where teams hit overlapping bugs already used by others. This is not a footnote. It is evidence that multiple researchers are independently finding the same categories of weaknesses, suggesting attackers can as well.

The Technical Breakdown Security Teams Should Care About

From a defensive lens, the most useful question is not "who won," but "what kinds of failures produced successful compromise on fully patched systems."

Recurring Vulnerability Classes

Weakness Type	Prevalence
Command injection paths	High
Buffer overflows (stack and heap)	High
Authentication weaknesses	Medium
Unsafe exposed methods/interfaces	High

Those weaknesses are amplified by the embedded reality of:

• Long-lived devices
• Mixed ownership boundaries
• Operational constraints that delay patch adoption

EV Charging: Beyond Just a Shell

EV charging stood out because of the add-on techniques and objectives that go beyond a shell. In multiple attempts, teams did not merely execute code - they demonstrated charging connector protocol and signal manipulation.

That is the escalation path defenders should model: a vulnerability chain that starts with a management interface or protocol weakness, then moves into control plane functions, and finally influences behavior that operators perceive as "equipment failure" rather than "security incident."

IVI Compromise Patterns

IVI systems sit at an uncomfortable intersection of consumer-facing complexity and automotive lifecycle constraints. The event included:

• Successful root-level outcomes against multiple head units
• USB-based attack against Tesla infotainment
• Collisions pointing to concentrated weakness hotspots (shared components, libraries, coding patterns)

Why This Is Bigger Than a Competition

The immediate "affected" population is the vendors and products that were targeted, but the broader impact lands on organizations that deploy and operate these technologies at scale.

EV Charging Operators

Chargers often sit on networks built for uptime and remote manageability rather than hostile threat environments. They may be connected via:

• Third-party maintenance channels
• Vendor cloud backends
• On-site connectivity that varies widely between locations

Automakers and Tier 1 Suppliers

The boundary between infotainment compromise and safety-critical domains is supposed to be controlled, but real systems contain:

• Diagnostic surfaces
• Bridging components
• Operational shortcuts

The right question: Does your architecture enforce separation under failure, not merely in design documentation?

Software Supply Chain Angle

Automotive Grade Linux and similar platforms represent the accelerating software supply chain reality of modern vehicles. A successful exploit chain against an automotive OS environment signals that secure development lifecycle controls need to meet enterprise software standards:

• Vulnerability intake processes
• Reproducible builds where possible
• SBOM-driven risk review
• Operational patch pipeline built for speed

How to Respond: The 90-Day Disclosure Window

The most actionable aspect of Pwn2Own Automotive 2026 is the disclosure cadence. Under ZDI's policy, vendors typically have a 90-day window to develop and release fixes before public disclosure.

For defenders, that window should be treated as preparation time, not passive waiting.

Step 1: Asset Certainty

Question	Why It Matters
Which EV charger models are deployed?	Cannot patch what you cannot find
Which firmware versions are present?	Determines vulnerability exposure
Which units are reachable from where?	Defines attack surface

Step 2: Plan for Patch Intake

• Pre-define maintenance windows
• Prepare rollback plans
• Establish monitoring signals for operational regression
• Build a staging pipeline that mirrors production

Step 3: Harden Assuming Future PoC

Control	Implementation
Network segmentation	Isolate EV charging from corporate IT and guest networks
Management access restriction	Known administration points only
Anomalous traffic monitoring	Watch charger management ports and protocols
Physical access review (IVI)	Enforce policy on USB/diagnostic processes

Why 76 Zero Days Changes the Conversation

Scale Matters

In earlier years, 49 zero days over three days was a meaningful indicator. Surpassing that by a wide margin indicates researchers found enough fertile ground to keep producing unique results throughout the event.

Collision Dynamics

Collisions mean two things simultaneously:

1. Defenders can prioritize certain weakness classes because independent teams converge on them
2. Attackers can likely reproduce those findings

The convergence is consistent with broader embedded security realities where unsafe interfaces, credential handling, and input validation failures recur across product lines.

Regulatory and Assurance Context

Automotive cybersecurity requirements are increasingly shaping supplier expectations and product governance. A record year at a public contest:

• Does not prove widespread wild exploitation
• Does raise the cost of inaction for organizations that must demonstrate due diligence

Boards and regulators will not be persuaded by "we run the latest version" when the public record shows fully patched systems still yielded meaningful compromise.

Closing

Pwn2Own Automotive 2026 will be remembered less for the prize total than for the record volume of demonstrated zero days on fully patched targets.

For most organizations, the correct response is neither panic nor dismissal. It is disciplined preparation during the coordinated disclosure window:

Action	Timeline
Inventory what you run	Now
Reduce exposure where you can	Now
Prepare to validate and deploy fixes	Before advisories go public
Plan maintenance without breaking operations	Within 90 days