🔍 Read the full analysis: PHP2Shell On Palo Alto Servers: Separating AI Assistance From Hype on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
Dream says an attacker used PHP2Shell and a separate Linux kernel flaw to gain root access on 24 Palo Alto Networks firewalls, then reached a government-affiliated organization’s internal firewall-management server. Palo Alto Networks says Unit 42 independently verified the tools and results. AI tools were found on the attacker’s servers, but Dream says there is no direct evidence they created the exploit or scripts.
Twenty-four Palo Alto Networks firewalls were compromised after an attacker chained the PHP2Shell flaws with a separate Linux kernel vulnerability, according to Israeli security company Dream. The activity, which Dream says began two days after PHP2Shell fixes were released, also led to access to a U.S. government-affiliated organization’s internal firewall-management server; Palo Alto Networks says its Unit 42 researchers independently verified the attacker’s tools and results.
Dream’s October 26 report, updated October 29, says the attacker ran an automated batch against 240 firewalls on the evening of October 23 and obtained root access on 24. Dream describes a chain involving PHP2Shell vulnerabilities CVE-2026-109618 and CVE-2026-109619, followed by Fragnesia, a Linux kernel privilege-escalation flaw tracked as CVE-2026-46300. The PHP flaws provided command execution as a restricted account, while the kernel bug was used to gain administrator-level access, according to Dream.
Dream says the attacker left hidden web shells on 22 of the compromised firewalls. Two belonged to the government-affiliated organization. After compromising a second firewall there on October 24, the attacker used a tunnel to reach the organization’s internal firewall-management server and then targeted its Active Directory system. Dream’s account identifies 25 compromised servers in total: 24 firewalls and the internal server. The targeted firewalls were in the United States, Japan, Taiwan and South Korea.
The timing of vendor fixes is central to the report. PHP released corrected versions on October 21, and Palo Alto Networks issued an advisory that day, but Dream says fixed PAN-OS builds for the affected 12.1 and 12.2 branches became available between October 27 and 29. Until a fix was available, Palo Alto advised keeping the management interface off the public internet. Dream says the attacker’s complete chain, including establishing a persistent root back door, took less than 15 minutes in the reported activity; that duration is Dream’s assessment.
Infographic 01 · PHP2Shell on Palo Alto firewalls
Two days from patch to root.
PHP fixed the bug on 21 October. Fixed firewall software for the attacked branches arrived almost a week later. The attack ran in between.
- 13–23 MayFragnesia goes public, kernel fix followsLinux privilege-escalation bug with a public exploit; CVE-2026-46300.
- 8 OctAttacker sets up an AI platformaiscan configured with DeepSeek and Kimi, per Dream’s timeline.
- 21 OctPHP fixes PHP2ShellPHP 8.5.12, 8.4.27, 8.3.36, 8.2.35. Palo Alto advisory the same day; PAN-OS fixes “upcoming”.
- 23 Oct, evening UTCBatch run: 240 targets, 24 rootedFound via device search engine, filtered to US, Japan, Taiwan, South Korea.
- 24 OctInto a government-affiliated networkTunnel, persistence, internal management server rooted (no. 25), Active Directory attacked.
- 26 OctDream publishes · CISA lists PHP2ShellFederal deadline 29 Oct.
- 27–29 OctFixed PAN-OS builds for 12.2 and 12.1CISA lists Fragnesia on 28 Oct; Unit 42 confirms Dream’s results on 29 Oct.
The fix for PHP existed. The fix for the firewall did not, yet. The open management page was the only door that mattered.Thorsten Meyer · my reading
Sources: Dream Research, 26 Oct 2026 (updated 29 Oct); php.net, 21 Oct 2026; Palo Alto Networks advisory (updated 29 Oct 2026); CISA KEV catalog, version 29 Oct 2026; oss-security, 13 May 2026; Unit 42, 29 Oct 2026. Day count: Thorsten Meyer.
Powered by Thorsten Meyer AI
Exposed Management Interfaces Raised Risk
The incident illustrates how internet-accessible management pages can turn a newly disclosed software flaw into a rapid compromise, especially when an attacker can pair it with an older, publicly known privilege-escalation vulnerability. A firewall is meant to control network access, but its management interface has powerful authority over security rules and connected systems. Compromise can therefore create a route beyond the device itself.
The reported movement into an organization’s internal firewall-management server and attempts against Active Directory matter because those systems help administer network defenses and user access. Dream’s findings describe access and targeting, not the full consequences: the source material does not establish that the attacker succeeded in taking control of Active Directory or what data, if any, was accessed or removed.
The report also offers a caution about the phrase “AI-assisted attack.” Dream found AI-related tools and configured models on the attacker’s servers, but says it has no direct evidence that AI wrote the exploit or scripts. The compromise is evidence of a fast, automated operation, not proof that an AI system independently planned or carried it out.
Products Worth Considering
Samsung 27 Odyssey OLED G5 (G50SF) QHD QD-OLED Gaming Monitor
LG 34GX90SA-W 34-inch Ultragear WQHD OLED Curved Gaming Monitor
DJI Mic 3 Wireless Microphone System (2 Transmitters, 1 Receiver, Charging Case)
As an affiliate, we earn on qualifying purchases.
Patches Followed the Disclosure
Dream says Fragnesia, the Linux kernel flaw used for privilege escalation, had been disclosed in May 2026 and had a public exploit. That made it a separate, previously known weakness rather than a newly discovered part of PHP2Shell. Chaining the two flaws allowed the attacker to move from limited command execution to root access, according to the report.
On October 21, the PHP project released updates for supported PHP branches and urged users to upgrade. Palo Alto Networks published its PAN-OS advisory on the same date, with fixed versions still forthcoming for the affected branches cited by Dream. The company’s guidance, as described in the source material, was to restrict access to the management interface rather than expose it to the internet while awaiting corrected builds.
Dream’s account says its researchers found roughly 19,000 PAN-OS firewalls with internet-facing management pages using FOFA, then obtained version information from about 5,000. Filtering by country and software version left 240 targets for the automated batch. These figures describe Dream’s scanning and selection process; they do not mean all 19,000 devices were vulnerable or attacked.
““No direct evidence” that AI tools wrote the exploit or scripts.”
— Dream
AI’s Role Remains Unverified
Dream reports finding Claude Code installed and DeepSeek and Kimi configured within an offensive-security platform called aiscan on the attacker’s servers. Those findings show that AI tools were present, but they do not establish how, or whether, the tools contributed to the intrusion. Dream explicitly says it has no direct evidence that they generated the exploit or scripts.
Other details also remain unresolved in the supplied account. It does not establish whether the attacker successfully accessed or altered Active Directory, what information may have been obtained, or whether every affected organization has completed remediation. The number of exposed devices is a scanning result, not a confirmed count of vulnerable systems. Palo Alto’s independent verification of tools and results does not, on its own, answer those questions.
Patch and Restrict Management Access
Organizations running affected PAN-OS branches should check Palo Alto Networks’ current advisory for the corrected version applicable to their systems, apply the available update, and keep management interfaces off the public internet. Operators of potentially affected devices should also investigate for compromise rather than treating installation of a patch as proof that no earlier intrusion occurred. The supplied report does not provide a complete set of indicators or a confirmed remediation status for each victim.
U.S. federal agencies were given a short remediation deadline after CISA added both PHP2Shell CVEs to its Known Exploited Vulnerabilities catalog on October 26, according to the source material. Further reporting and vendor or government notices may clarify the scope of affected systems, the outcome of the Active Directory targeting, and any confirmed use of AI tools in the operation. Until then, the established account is an automated exploit chain with AI-related software present—not a demonstrated AI-written attack.
Key Questions
What does PHP2Shell refer to in this report?
PHP2Shell is the name used for two vulnerabilities, CVE-2026-109618 and CVE-2026-109619. Dream says they were chained with the separate Fragnesia Linux kernel flaw to gain root access on targeted firewalls.
How many systems did Dream say were compromised?
Dream reported root access on 24 firewalls and access to one internal firewall-management server, for 25 servers in total. It said the firewalls were in the United States, Japan, Taiwan and South Korea.
Did AI write or run the attack?
That has not been established. Dream found AI tools and configured models on the attacker’s servers but said it had no direct evidence that they wrote the exploit or scripts, and that the operation did not appear fully autonomous.
What should organizations with affected firewalls do?
Check Palo Alto Networks’ advisory and install the corrected PAN-OS build for the relevant branch. Management interfaces should not be exposed to the public internet, and potentially affected devices should be checked for signs of prior compromise.
Source: ThorstenMeyerAI.com
