Printers get hacked because they are networked computers that most people never treat as computers. Attackers log in to the admin web page with a factory default password, reach print ports that were exposed to the internet by mistake, and exploit firmware bugs in the code that parses fonts, PDFs, images and network requests. They also go after the computers that print, through the Windows print spooler and the CUPS printing system, and after what the printer remembers: stored scans, queued jobs and saved passwords for email, file shares and the company directory.
The consequences are real. In 2019 Microsoft reported that attackers had used an office printer, a phone and a video decoder as entry points into corporate networks, in two cases because the default password had never been changed. In June 2025 Rapid7 disclosed that the default admin password on 695 printer models could be worked out from the device's serial number, a flaw the vendor said could not be fully fixed in firmware.
This guide explains each attack from the attacker's point of view so that you can defend against it: how it works, what the attacker needs, documented cases with CVE numbers, the impact, how to spot it and what to do. It ends with hardening checklists for homes, offices and fleet managers, and with the part most guides skip: the firmware and software supply chain inside the printer, which is the real reason so many stay vulnerable for years.
It is written for home users with an inkjet on the desk, for IT teams running multifunction devices in offices, for fleet managers and managed print providers responsible for hundreds of devices, and for the engineers who build printers and their software. Jargon is defined on first use.
This is a defender's guide. It contains no exploit code, no tool commands and no step-by-step attack instructions. Where a security tool is mentioned, it is only as context for authorised testing. Testing a printer or network that you do not own, or have no written permission to test, is illegal in most countries. If you want to test your own equipment, do it on hardware you own, on a network you control, or under a signed scope of work.
The printer is only as safe as the network around it. The router in front of it is covered in How routers get hacked and how to secure them, the wireless link in How Wi-Fi gets hacked and how to secure it, and the cameras, speakers and plugs that share its network in How smart home devices get hacked and how to secure them.
Contents
- How does a printer work, and what software does it run?
- What is a printer's attack surface?
- Which printer attacks matter most?
- How do default passwords and exposed web panels get printers hacked?
- How do firmware vulnerabilities let attackers take over a printer?
- What did Pwn2Own reveal about printer security?
- What is PrintNightmare, and why is the Windows print spooler risky?
- What were the 2024 CUPS vulnerabilities?
- What can attackers get from stored scans and print jobs?
- How do attackers use a printer as a network foothold?
- How can a fax line be used to attack a network?
- Why do printer fleets stay unpatched?
- How can you tell if a printer has been hacked?
- How do you secure a home printer?
- How do you secure printers in an office or a managed fleet?
- What is in the printer firmware supply chain?
- What should printer makers do?
- How Safeguard helps printer and MFP makers
- The takeaway
- Frequently asked questions
- Sources
How does a printer work, and what software does it run?
A network printer is an embedded computer with an operating system, a network stack, a web server and several interpreters that turn incoming data into pages. A multifunction printer, or MFP, adds scanning, copying, faxing, an address book and often a hard drive. All of it is firmware, and every part that reads outside input can contain bugs.
When you press Print, your computer hands the document to its print system. On Windows that is the Print Spooler service; on Linux, Unix and ChromeOS it is usually CUPS, the Common Unix Printing System. The print system converts the document into a page description language the printer understands, and sends it over the network or USB.
Three ways of sending jobs dominate. IPP, the Internet Printing Protocol, runs over port 631 and, in its encrypted form IPPS, over HTTPS. LPD, the old line printer daemon protocol, still exists on many devices. Raw printing on TCP port 9100, often called JetDirect after HP's name for it, simply accepts whatever bytes arrive and feeds them to the interpreter.
The printer then has to read those bytes. A page description language is a small programming language for pages. PCL (Printer Command Language) and PostScript are the classics, PDF is now common, and PJL (Printer Job Language) wraps jobs with control commands such as which tray to use. PostScript is a full programming language, and PJL can read and change device settings. That matters, because anything that can send a print job can talk to these interpreters.
Around the printing core sit more services. An embedded web server serves the admin page. SNMP (Simple Network Management Protocol) reports status and toner levels to management tools. Discovery protocols such as mDNS (Bonjour), WSD (Web Services for Devices) and SLP (Service Location Protocol) announce the printer to computers nearby. Office MFPs add scan to email (using SMTP), scan to folder (using SMB or FTP), an address book, LDAP lookups against the company directory so users can find colleagues' addresses, a fax modem connected to a phone line, and increasingly a vendor cloud service and a phone app.
Figure 1. A multifunction printer is a networked computer with a paper path attached.
What is inside printer firmware?
Underneath it all is an embedded operating system. Some printers run Linux; others run a commercial real-time operating system (an RTOS) or the vendor's own OS. The network stack may be licensed from a third party. In 2020 the Ripple20 research found 19 vulnerabilities in a TCP/IP library from Treck that had spread through device makers for about two decades, and JSOF's demonstration used an HP OfficeJet Pro 8720 printer. We cover the Treck stack in more detail in CVE-2020-11899 and the Treck TCP/IP stack.
Font engines, image decoders and PDF interpreters may be licensed from specialist vendors, taken from open source, or written in house. Some are old. When Check Point examined HP OfficeJet fax handling in 2018, the vulnerable code was image parsing for incoming faxes.
Storage varies widely. A home inkjet usually has only flash memory. Office MFPs often carry a hard drive or SSD that holds the print queue, retained jobs, scanned images, the address book and logs. That drive outlives the job by years, and sometimes outlives the printer.
Home printers and office MFPs are different risks
A home printer usually has fewer services, no hard drive and a single user. Its main risks are a default admin password, firmware that never gets updated, and being reachable from the internet by accident. An office MFP has everything: dozens of users, stored credentials for other systems, a fax line, a drive full of documents, and often a service contract under which someone else manages it. The attacks below apply to both, but the office device has far more to lose.
What is a printer's attack surface?
A printer can be attacked from six directions: from the internet if it is exposed, from inside the network through its admin page, print ports and protocols, through a phone line if it faxes, in person through its panel and USB ports, through the computers and servers that print to it, and through its supply chain.
The attack surface is every place an attacker can send input to a system. For a printer it is much larger than people expect.
Figure 2. Six directions of attack, and the most often forgotten is the computer that sends the job.
From the internet. A printer should never be directly reachable from the internet, yet many are. In 2020 CyberNews researchers found more than 800,000 internet-accessible printers with network printing enabled using search engines that index internet-connected devices. Common causes are a printer given a public IP address, a port forward set up years ago for remote printing, or a router with UPnP (Universal Plug and Play) opening ports automatically, which the router guide explains in full.
From inside the network. Any device on the same network, such as a laptop infected by phishing or a compromised smart TV, can usually reach the admin page, print ports and management protocols. The 2017 academic study SoK: Exploiting Network Printers from Ruhr University Bochum tested 20 printer models and found all of them vulnerable to at least one attack through PostScript and PJL, ranging from denial of service to extracting print jobs and system files.
From a browser. The same study showed how cross-site printing lets a malicious web page make a visitor's browser send data to a printer on their local network, so an attacker on the internet can reach a printer they cannot connect to directly. The HP Printing Shellz research in 2021 used the same idea, and CSO Online described it as the most dangerous vector because it enables attacks from outside the LAN.
Through the phone line. A fax-capable printer accepts data from anyone who dials its number. Fax runs over the telephone network, not the internet, so a firewall never sees it.
In person. The touch panel, front USB ports, and in some cases internal debug or service ports. HP's CVE-2021-39237 involved exposed ports on the MFP board that gave privileged access with about five minutes of physical access.
Through the host print stack. The Windows Print Spooler and CUPS run on almost every computer. They talk to printers, load drivers and parse what printers report back. PrintNightmare and the 2024 CUPS chain both lived on this side.
Through the supply chain. The firmware update channel, the signing keys, third-party components inside the firmware, fleet management tools and the vendor's cloud services. A weakness here reaches every device of a model at once.
Which printer attacks matter most?
The attacks that matter most are the cheap ones: default or weak admin passwords, known firmware bugs left unpatched, printers exposed to the internet, and stored credentials that leak to anyone with admin access. Parser zero-days, fax exploits and malicious firmware are real but need more skill. Fix the cheap ones first.
Attackers do not usually start with a zero-day. They start with a scan for open ports and a list of default passwords, then try exploits for flaws that were patched months ago, because many printers never receive the patch. Figure 3 ranks the attack classes in this guide by how much effort they cost the attacker and how much damage they do.
Figure 3. Start in the top-left box: cheap for attackers, severe for you.
The history shows the same pattern. Research and incidents have repeated the same lessons for more than a decade: printers hold sensitive data, accept unsigned or weakly checked input, sit on flat networks and are patched late.
Figure 4. Fifteen years of printer security, with the same lessons repeated.
How do default passwords and exposed web panels get printers hacked?
Many printers ship with a blank, well-known or predictable admin password, and their admin web page and print ports are reachable by anyone on the network, or on the internet if exposed. An attacker who logs in can change settings, read stored data and redirect scans. It is the cheapest printer attack and still the most common.
How it works, and why it works
Every network printer has an admin interface, normally the embedded web server (EWS) you reach by typing the printer's IP address into a browser. From there an administrator can change network settings, scan destinations, LDAP and email accounts, security options and firmware. Many models also accept configuration through SNMP and PJL.
For years, many printers shipped with no admin password at all, or with the same password on every unit, printed in the manual. Newer devices often use a unique per-device password printed on a label. That is better, but only if the password is truly random. The Brother case below shows what happens when it is not.
Raw print ports are a related problem. Port 9100 accepts anything sent to it, without authentication, on most devices. On a printer exposed to the internet, that means anyone in the world can print to it and talk to its interpreters.
What the attacker needs
Network reach to the printer, either from inside the network or from the internet if it is exposed, and a password that is blank, published or derivable. For internet-exposed devices, search engines that index connected devices list them by protocol and manufacturer, so locating them takes no skill.
Real cases
About 50,000 printers forced to print a message (2018). In late 2018 a person calling himself TheHackerGiraffe searched for internet-exposed printers using the IPP, LPD and JetDirect protocols, found up to 800,000, and sent a message to 50,000 printers on port 9100 urging readers to subscribe to the YouTuber PewDiePie. He used the Printer Exploitation Toolkit (PRET), an open source tool released by the Bochum researchers for authorised printer security testing. No passwords were needed: the raw print port accepted the job.
27,944 printers hijacked to print a security guide (2020). CyberNews researchers repeated the exercise as an awareness test. From the 800,000 internet-accessible printers they found, they sampled 50,000 and successfully made 27,944 of them print a short guide on securing printers, a 56 percent success rate. Their advice: restrict printer ports, block unused protocols with a firewall, update firmware, and change default administrator credentials.
An office printer as a way into corporate networks (2019). Microsoft's Security Response Center reported that an activity group had used a VoIP phone, an office printer and a video decoder to get into several customers' networks. In two cases the default manufacturer passwords had not been changed, and in the third the latest security update had not been applied. Once inside, the attackers scanned the network, captured traffic on local subnets and looked for higher-privileged accounts.
Brother default passwords derived from serial numbers (2025). Rapid7 disclosed eight vulnerabilities affecting 748 models across five vendors: 689 Brother models plus 46 from FUJIFILM Business Innovation, six from Konica Minolta, five from Ricoh and two from Toshiba Tec. The most serious, CVE-2024-51978 (CVSS 9.8), affects 695 models: the default administrator password is generated from the device's serial number during manufacturing, so anyone who learns the serial number can calculate the password. A second flaw, CVE-2024-51977, leaks the serial number to unauthenticated requests over HTTP, HTTPS and IPP. Brother told Rapid7 that CVE-2024-51978 cannot be fully remediated in firmware; only devices made under a changed manufacturing process are fully fixed, and existing units need a workaround. SecurityWeek reported that millions of enterprise and home printers are believed to be exposed.
Impact
With admin access an attacker can change DNS and network settings, redirect scan to email or scan to folder destinations, read the address book, read or replace stored LDAP, SMTP, SMB and FTP credentials (see the pass-back section below), turn on services such as FTP or Telnet, and on some models install firmware or apps. With only print-port access they can waste paper and toner, cause denial of service, and on vulnerable devices reach the interpreters where code execution bugs live.
How to detect it
- Check the admin password status on every device. Many fleet tools report whether the default is still set.
- Scan your own public IP ranges for ports 631, 9100, 515 (LPD), 80 and 443 answering as a printer. If any respond, the printer is exposed.
- Watch printer logs, where available, for admin logins from unexpected addresses and for changes to scan destinations, LDAP servers and DNS settings.
- Unexpected print jobs, especially odd messages or pages of garbage, are a classic sign that a print port is reachable from somewhere it should not be.
How to defend against it
- Change the admin password to a long, unique one on every device, even if it shipped with a label password. For Brother and the other affected brands, Rapid7 and the vendors advise exactly this, because the default cannot be trusted.
- Keep printers off the internet. No public IP, no port forwards, UPnP off on the router. If remote printing is needed, use the vendor's authenticated cloud service or a VPN.
- Restrict who can reach the admin page. Allow admin access only from a management network or named admin workstations, and turn on HTTPS for the admin interface.
- Require authentication for printing where the device supports it, and disable raw port 9100 printing if your environment does not need it.
- Use named admin accounts and directory integration on office fleets, so changes are attributable.
How do firmware vulnerabilities let attackers take over a printer?
Printer firmware parses untrusted input all day: fonts, PDFs, PostScript, images, fax data and network requests. Bugs in that parsing code, mostly memory-safety errors such as buffer overflows, let an attacker run code on the printer. HP, Canon, Lexmark, Xerox and Brother have all fixed critical flaws of this kind in recent years.
How it works, and why it works
Printer firmware is mostly written in C and C++, languages where a missing length check can let input overwrite memory. A buffer overflow happens when code copies more data into a fixed-size area than it can hold. A type confusion happens when code treats data as the wrong kind of object. Either can let carefully shaped input take control of the program. Printers make this worse in three ways: they parse complex formats from anyone who can print, they often run everything with full privileges, and older models lack the memory protections common on PCs and phones.
A second class of firmware flaw is about logic rather than memory: server-side request forgery (SSRF), where the attacker makes the printer send requests on their behalf, command injection, where input ends up in a system command, and authentication bypass.
What the attacker needs
Usually just the ability to send data the vulnerable code will parse: a print job, a request to a web service, a document a user prints, or a fax. Writing a working exploit takes skill, but once a proof of concept is public, reuse takes much less.
Real cases
HP: Printing Shellz, CVE-2021-39238 and CVE-2021-39237 (2021). F-Secure researchers found two flaws affecting more than 150 HP multifunction models running FutureSmart firmware, dating back to at least 2013. CVE-2021-39238 was a font parsing buffer overflow that could lead to code execution and was described as wormable, meaning an infected printer could spread the attack to other vulnerable printers on the network. Routes in included printing a booby-trapped PDF, a web page using cross-site printing, and printing from another device on the same segment. CVE-2021-39237 involved exposed debug ports reachable with physical access. HP issued patches; the researchers advised updating, isolating printers in their own firewalled segment, routing printing through a print server and restricting printer-to-printer traffic.
Canon: seven critical buffer overflows (2024). In February 2024 Canon fixed CVE-2023-6229 through CVE-2023-6234 and CVE-2024-0244, all rated CVSS 9.8 by Japan's JVN portal. They sat in processes such as PDL resource download, fax number handling, address book credentials, WSD probe requests and SLP attribute requests, and could be exploited over the network for code execution or to make the device unresponsive. Affected small office models included imageCLASS, i-SENSYS and Satera lines on firmware 03.07 and earlier. All seven were reported through Trend Micro's Zero Day Initiative. Canon advised updating firmware, giving printers private IP addresses behind a firewall, and never connecting them directly to the internet.
Lexmark: CVE-2023-23560 (2023). An SSRF flaw in the Web Services feature of more than 100 Lexmark models, rated 9.0 by Lexmark, which said it could allow arbitrary code execution. Firmware 081.233 and earlier was affected; the fix shipped in later releases. A researcher published proof of concept code. Lexmark said it was not aware of malicious use, and offered a workaround of disabling the Web Services port, TCP 65002, where patching was not possible.
Xerox: pass-back flaws CVE-2024-12510 and CVE-2024-12511 (2025). On Xerox VersaLink MFPs with firmware 57.69.91 and earlier, someone with admin access could redirect LDAP lookups or SMB and FTP scan destinations to their own server and capture the stored credentials. The stored data section below explains this pass-back pattern.
Brother and four other vendors (2025). Beyond the default password flaw, Rapid7's eight CVEs included a stack buffer overflow, two SSRF flaws, two denial of service flaws and an LDAP and FTP pass-back flaw. Seven of the eight were fixed in firmware.
HP: unsigned firmware updates (2011). Columbia University researchers showed that several HP LaserJet models did not require firmware updates to be digitally signed, so modified firmware could be loaded through the remote firmware update process by anyone with network access to it. HP disputed parts of the research, said a hardware thermal breaker prevented the fire risk the researchers raised, and worked on a fix. In 2015 HP added firmware protections to its enterprise LaserJets, including BIOS integrity checks at boot, a check that firmware is signed by HP before it loads, and run-time intrusion detection. That is the direction every maker should take.
Impact
Code execution on a printer gives the attacker everything the printer has: stored jobs and scans, credentials, network access, and a persistent hiding place that few security teams monitor. Wormable flaws such as CVE-2021-39238 can spread to every vulnerable printer on the network. Denial of service flaws stop printing, which matters in hospitals, warehouses and shops that run on printed labels and forms.
How to detect it
- Keep a firmware version inventory for every model and compare it with each vendor's security advisories. This is the only reliable way to know which devices carry known vulnerabilities.
- Watch for printers that reboot unexpectedly, change configuration on their own, or start making outbound connections to unfamiliar addresses.
- Use network monitoring to flag printers talking to anything other than print servers, management tools and the vendor's update service.
How to defend against it
- Patch firmware on a schedule, with a fast track for critical and exploited flaws. Turn on automatic updates where the vendor offers them.
- Turn off features you do not use, especially web services, WSD, SLP, FTP and Telnet. Several of the flaws above were in optional services; Lexmark's own workaround was to disable one.
- Segment printers so only print servers and named management hosts can reach them.
- Print through a print server, so user devices never talk to printers directly, which blocks cross-site printing and printer-to-printer spread.
- Buy devices with signed firmware and secure boot, and replace models that no longer receive updates.
What did Pwn2Own reveal about printer security?
At Pwn2Own, the Zero Day Initiative's hacking contest, researchers took over Canon, HP and Lexmark printers at every event reviewed here, in 2022, 2024 and 2025, almost always with memory-safety bugs such as buffer overflows. Each successful entry is a previously unknown vulnerability that the vendor then fixes.
Pwn2Own pays researchers for working exploits against fully patched devices, then hands the vulnerabilities to vendors to fix. Printers are a regular category at the contest's consumer and small office events. This guide reviews Pwn2Own Toronto in December 2022 and Pwn2Own Ireland in 2024 and 2025.
On day one in Toronto, teams took over a Canon imageCLASS MF743Cdw, a Lexmark MC3224i and an HP Color LaserJet Pro M479fdw using stack buffer overflows and command injection. The contest also introduced the SOHO Smashup, in which a team must first break into a router from the internet and then pivot to a device behind it. On the first day DEVCORE went through a MikroTik router to a Canon printer, and Neodyme went through a NETGEAR router to an HP printer.
At Pwn2Own Ireland in October 2024, Team Neodyme exploited an HP Color LaserJet Pro MFP 3301fdw with a stack buffer overflow and another team took a Canon imageCLASS MF656Cdw. On day three, Viettel Cyber Security used a type confusion bug on a Lexmark CX331adwe, and the PHP Hooligans / Midnight Blue team completed a SOHO Smashup from a QNAP router to a Lexmark printer.
In October 2025, Neodyme exploited an HP DeskJet 2855e and four separate teams exploited the Canon imageCLASS MF654Cdw on day one alone; more Canon and Lexmark entries followed on day two.
Figure 5. 15 of the 17 printer entries with a named bug class were memory-safety bugs.
We counted the bug classes in the 20 printer entries on the five ZDI results pages listed in the sources. Fifteen of the 17 entries where ZDI named the bug class were memory-safety bugs: stack and heap overflows, out-of-bounds writes, type confusion and a bad pointer release. That is a count from those pages, not every day of every contest, but the pattern is clear. Printer firmware is still largely written in memory-unsafe code that parses network input, and skilled researchers find exploitable bugs in current models within months of the devices being chosen.
Two lessons follow. For buyers: a fully patched printer is not invulnerable, so segmentation and least privilege matter as much as patching. For makers: the bugs are concentrated in parsers and network services, which is exactly where fuzzing, static analysis, memory-safe rewrites and dependency tracking pay off.
What is PrintNightmare, and why is the Windows print spooler risky?
PrintNightmare (CVE-2021-34527) was a 2021 remote code execution flaw in the Windows Print Spooler, the service that manages printing on almost every Windows machine. It let an authenticated user run code as SYSTEM, including on domain controllers. Public exploits appeared quickly, and the fix required both patches and configuration changes.
How it works, and why it works
The Print Spooler is a Windows service that queues jobs, talks to printers and print servers, and installs printer drivers. It runs as SYSTEM, the highest privilege on a Windows machine, it is enabled by default, and it is reachable over the network on many systems. A printer driver is code, so any path that lets a user make the spooler load a driver of their choosing is a path to running code with SYSTEM rights.
Point and Print is a Windows feature that lets users connect to a shared printer and automatically download its driver from the print server. It is convenient, and it is exactly the kind of feature attackers look for.
In June 2021 Microsoft patched a spooler flaw, CVE-2021-1675. Researchers then published exploit code they believed targeted that flaw, and it worked on patched systems. Microsoft confirmed the researchers had found a separate vulnerability, which became CVE-2021-34527, PrintNightmare. Rapid7 noted that all versions of Windows were vulnerable by default, not only domain controllers, and that public exploits still achieved code execution on patched systems whenever Point and Print was left in an insecure configuration.
An earlier flaw shows the same theme. PrintDemon (CVE-2020-1048), patched in May 2020, let a standard user create a printer port that pointed at a protected file so that the spooler, running as SYSTEM, would write print data into it. The researchers who described it noted that a malicious port created before patching could remain afterwards, so they advised auditing existing ports as well as patching.
What the attacker needs
For PrintNightmare: valid credentials on the domain or machine (any ordinary user account would do) and network access to a system running the spooler. For PrintDemon: a local user account.
Impact
Code execution as SYSTEM, which on a domain controller means control of the whole Windows domain. Rapid7 reported exploitation in the wild and warned the flaw could be used in ransomware campaigns.
How to detect it
- Confirm that every Windows system has the July 2021 and later cumulative updates.
- Check the Point and Print registry settings Microsoft documents, and audit which systems actually need the spooler.
- Rapid7 suggested reviewing the Microsoft-Windows-PrintService/Admin event log for messages that the spooler failed to load a plug-in module, which can indicate exploitation attempts.
How to defend against it
- Patch every Windows system and keep it patched, because the spooler is old, privileged code that will keep needing fixes.
- Disable the Print Spooler on systems that do not print, starting with domain controllers and servers. Rapid7 recommended exactly this.
- Restrict Point and Print so users can only install drivers from approved print servers, and keep the elevation prompts on.
- Turn on Windows protected print mode on Windows 11 where your printers support it. It uses only Microsoft's modern print stack and removes printers that rely on third-party drivers. Microsoft says it works with Mopria certified printers, and that over 120 million printers already sold carry that certification.
- Plan for driverless printing. Under Microsoft's end of servicing plan for third-party printer drivers, Windows prefers its built-in IPP class driver from 1 July 2026, and from 1 July 2027 only security fixes to existing third-party drivers will be allowed. Fewer third-party drivers means less code running inside the spooler.
For a deeper technical walk-through, see our post PrintNightmare, CVE-2021-34527, explained.
Figure 6. The computer that prints is part of the attack surface.
What were the 2024 CUPS vulnerabilities?
In September 2024 researcher Simone Margaritelli disclosed four chained vulnerabilities in CUPS, the printing system on most Linux and Unix machines. CVE-2024-47176 in cups-browsed let a remote attacker add a fake printer; when a user printed to it, CVE-2024-47177 in cups-filters let the attacker's commands run.
How it works, and why it works
CUPS is the print system on most Linux distributions, the BSDs, ChromeOS and other Unix-like systems. A helper called cups-browsed listens for printers announced on the network and adds them automatically, which is convenient on an office network full of printers.
Margaritelli's write-up described four flaws that chain together:
- CVE-2024-47176 (cups-browsed 2.0.1 and earlier): the service listened on UDP port 631 on all network interfaces and trusted packets from any source, so an attacker could make it fetch printer details from an address the attacker chose.
- CVE-2024-47076 (libcupsfilters): the printer attributes returned by that server were not validated.
- CVE-2024-47175 (libppd): those unvalidated attributes were written into a printer description file, letting the attacker inject directives.
- CVE-2024-47177 (cups-filters): a filter called foomatic-rip could be made to run arbitrary commands through one of those injected directives.
The net effect: a remote attacker could add a malicious printer to a target system, and the next time someone printed to it, the attacker's commands ran on that system. Elastic Security Labs noted that exploitation requires user interaction, because someone must start a print job to the fake printer, and that a phishing document could be used to prompt that.
What the attacker needs
The ability to send a UDP packet to port 631 on the target, either across the internet if the port is exposed or from the same local network, where spoofed printer announcements can also work. Then a user who prints to the new printer.
Real-world exposure
Margaritelli reported seeing hundreds of thousands of systems connect back during his internet-wide scans. Elastic cited a count of over 75,000 CUPS services exposed on the internet. Akamai's researchers found more than 198,000 vulnerable devices reachable from the internet, and showed that tens of thousands could be abused for DDoS amplification: a 30-byte packet caused an average of about 18,000 bytes of traffic toward a victim, roughly 600 times amplification.
How to detect it
- Find every system with cups-browsed installed and running, and check CUPS package versions against your distribution's advisories.
- Look for printers that appeared on a system without anyone adding them.
- Elastic published detection rules for shell commands spawned by foomatic-rip or cupsd, shells run by the lp user, and network connections from foomatic-rip child processes.
- Akamai suggested filtering on suspicious IPP requests to identify amplification traffic.
How to defend against it
- Update CUPS packages on every Linux and Unix system.
- Remove or disable cups-browsed where automatic printer discovery is not needed, which is most servers.
- Block UDP 631 at the network edge and from untrusted networks; internet-facing servers should never accept it.
- Treat printing on servers as an exception. A database server does not need a print stack.
Our deep dive on CVE-2024-47176 covers the chain and its fixes in more detail.
What can attackers get from stored scans and print jobs?
Office printers keep more than people think: queued and retained print jobs, scanned images, fax logs, an address book, and saved passwords for the email server, file shares and company directory. An attacker with admin access can harvest those credentials through a pass-back attack, and anyone who gets the drive from a disposed device can read old documents.
Figure 7. What a printer remembers can outlive the job by years.
Documents on the drive
Many office MFPs write jobs and scans to internal storage as part of normal operation, and some keep them for features such as reprinting or secure pull printing, where a job waits until the user badges in at the device.
The risk was made public in April 2010, when CBS News bought used copiers, four of them according to Legaltech News, and worked with a copier security expert to examine their drives. TruthOrFiction's summary of the report says the copiers cost about $300 each and that medical records, Social Security numbers and confidential police records were recovered. The same summary noted that Xerox said it had included an overwrite feature that erases stored image data since 2008. Features like that now exist across vendors, but they only help if they are turned on and the device is wiped before it leaves.
The problem has not gone away. HP's Wolf Security survey of more than 800 IT and security decision makers, reported in July 2025, found that 86 percent see data security as a barrier to reusing, reselling or recycling printers, and 35 percent are unsure printers can be fully and safely wiped. HP sells printers, so read its research with that in mind, but the numbers match what IT teams report.
Home printing is part of the same story. Quocirca's 2025 survey of 400 IT decision makers found that 56 percent had at least one print-related data loss in the previous year, and that documents printed on home printers were a top factor in data loss for 53 percent.
Pass-back: harvesting stored credentials
To scan to a folder, look up email addresses or send scans by email, an MFP stores credentials for other systems. That makes it a credential store with a web page in front of it.
A pass-back attack works like this, in concept. The printer holds a saved password for, say, the LDAP directory, but it will not display it. An attacker with admin access to the printer changes the server address so the printer connects to a machine the attacker controls, then triggers a test or lookup. The printer dutifully sends its stored credentials to the wrong server. Rapid7's Xerox advisory documented exactly this for LDAP, SMB and FTP, and its Brother research found the same class of flaw for LDAP and FTP in CVE-2024-51984.
Why it matters: the service account behind scan to folder often has write access to large file shares, and LDAP accounts are directory accounts. Rapid7 specifically advised against using privileged accounts, such as domain admin accounts, for LDAP or scan-to-file services. That advice exists because people do.
What the attacker needs
For pass-back, admin access to the printer, which a default password provides. For stored documents, either code execution on the device, a vulnerable job retrieval feature, or physical access to the drive, most often after disposal.
How to detect it
- Alert on configuration changes to LDAP servers, SMTP relays and scan destinations.
- Watch directory logs for the printer's service account authenticating from a machine that is not the printer.
- Audit which privileges scan and LDAP accounts actually have.
- Track every device with storage through disposal, with a wipe record per serial number.
How to defend against it
- Use least-privilege service accounts for scan to email, scan to folder and LDAP lookups, each limited to exactly what the feature needs. Never use admin accounts.
- Lock the admin plane with unique passwords and named accounts, which blocks most pass-back attempts at the start.
- Turn on storage encryption and automatic overwrite where the device supports it, and keep job retention short.
- Use pull printing for sensitive work so jobs print only when the user is standing at the device.
- Wipe or destroy storage before disposal, with a certificate for each device, and write that duty into leasing and service contracts.
How do attackers use a printer as a network foothold?
A compromised printer is an ideal hiding place: it is always on, rarely monitored, trusted on the network, and often holds credentials. Attackers use it to capture traffic, scan the network, harvest stored passwords and move to servers and other printers. Microsoft, Check Point, F-Secure and Pwn2Own have all documented the pattern.
A foothold is an attacker's first controlled position inside a network, from which they explore and move further, which is called lateral movement.
Figure 8. How a printer becomes a foothold, and the defence at each stage.
How it works, and why it works
Printers sit on the same flat network as staff laptops in many offices. They are trusted to talk to file servers, mail servers and the directory. Endpoint security tools that protect PCs usually do not run on them, and their logs rarely reach a security team. An attacker who controls one can stay for a long time.
Real cases
- Microsoft, 2019. After getting in through a printer, a phone and a video decoder, the attackers scanned for other insecure devices, captured traffic on local subnets and looked for administrative accounts, and dropped scripts on the devices so they would persist.
- Faxploit, 2018. After taking over an HP OfficeJet through its fax line, Check Point's researchers demonstrated spreading through the connected network using the EternalBlue exploit for unpatched Windows machines.
- Printing Shellz, 2021. F-Secure described CVE-2021-39238 as wormable, able to spread from one vulnerable MFP to others on the same network.
- Pwn2Own SOHO Smashup, 2022 and 2024. Teams broke into a router from the internet and then used it to reach and take over a printer behind it. The path works in both directions: a router compromise exposes printers, and a printer compromise exposes everything the printer can reach.
How to detect it
- Baseline printer network behaviour. A printer normally talks to print servers, management tools, mail and file servers it scans to, and its vendor's update service. Anything else, especially outbound connections to the internet or scanning of internal ranges, is a red flag.
- Send printer logs to your SIEM (security information and event management system) where the device supports syslog.
- Use network detection to flag printers authenticating to servers they have never used before.
How to defend against it
- Put printers on their own VLAN (a virtual network segment) with firewall rules that allow only print servers and management hosts in, and only the services they need out.
- Block printer access to the internet except the vendor's update and cloud endpoints, if you use them.
- Use least-privilege service accounts, as above, so a stolen printer credential opens as little as possible.
- Treat printers as assets in your security program: inventory, vulnerability management, logging and incident response. Microsoft's 2019 recommendations included cataloguing every IoT device, using a separate network for them, auditing configuration and patches regularly, and including IoT intrusion scenarios in red team testing.
How can a fax line be used to attack a network?
In 2018 Check Point showed that a fax-enabled HP OfficeJet could be taken over by sending it a specially crafted fax, and then used to attack the rest of the network. The attack, Faxploit, needed only the fax number and travelled over the phone line, where no firewall looks. HP fixed the flaws as CVE-2018-5924 and CVE-2018-5925.
How it works, and why it works
A fax is just data arriving over a phone line, and the printer's firmware has to decode it. In Faxploit, Check Point researchers Yaniv Balmas and Eyal Itkin found two flaws in how HP OfficeJet all-in-ones parsed incoming fax images: CVE-2018-5925, a buffer overflow when parsing COM markers, and CVE-2018-5924, a stack-based flaw when parsing DHT markers. COM and DHT markers are structures inside JPEG image data. A specially crafted image sent by fax could take control of the printer.
The insight was about reach, not just the bug. Check Point pointed out that because the attack uses the telephone line, a network does not even need an internet connection to be reached this way. The fax modem bypasses every firewall, email filter and web proxy in the building.
What the attacker needs
The target's fax number, which is often printed on letterheads and websites, and a working exploit for the specific firmware.
Real cases and scale
Check Point demonstrated the attack on HP OfficeJet Pro all-in-ones and presented it in August 2018. It said the attack had not been seen in the wild, and HP released patches, delivered automatically to most customers. Check Point cited around 46.3 million fax machines in use, 17 million of them in the US, and noted that other vendors' devices and fax services could be affected because the weakness class lies in how fax data is handled.
Impact
Full control of the printer, theft or alteration of incoming faxes, and a foothold inside the network that bypassed perimeter defences entirely.
How to detect it
- Know which devices have an active phone line plugged in. Many offices have forgotten fax lines on MFPs.
- Treat fax-capable devices as internet-facing for patching priority, since anyone with a phone can send them data.
How to defend against it
Check Point's own advice is the right list:
- Disconnect the phone line if you no longer use fax. Most organisations can.
- Patch fax-capable devices first, alongside anything internet-facing.
- Segment them so that a compromised fax printer cannot reach the rest of the network.
- Consider a fax server or cloud fax service instead of fax modems on every MFP, so one hardened system handles inbound fax.
Why do printer fleets stay unpatched?
Printer firmware is patched far less reliably than PCs: an HP survey published in 2025 found only 36 percent of IT teams promptly apply printer firmware updates. Fleets mix brands and ages, updates are manual or vendor-managed, devices stay in service for years, and some flaws, like Brother's default password issue, cannot be fixed in firmware at all.
The numbers tell the story. In HP Wolf Security's survey, only 36 percent of IT teams promptly applied printer firmware updates, just 35 percent could identify which printers were vulnerable when new flaws were disclosed, 51 percent could not confirm whether printers had been tampered with during delivery, and IT staff spent an average of 3.5 hours per printer each month on printer security. The average organisation had around 80 idle or near end-of-life printers.
Several causes compound each other:
- Mixed fleets. Quocirca found 59 percent of organisations run multi-vendor fleets, and that maintaining patches and updates across them is an administrative burden. Each vendor has its own tools, schedules and advisory formats. Quocirca put the average cost of a print-related breach at £937,000 for multi-vendor fleets against £630,000 for standardised ones.
- Manual or outsourced updates. Many office devices are leased and serviced by a dealer or managed print provider. Firmware updates may wait for a service visit, or for nobody at all, because neither side thinks it is their job.
- Fear of breakage. Updates can change behaviour, break integrations with document systems, or require a reboot during business hours. So they are deferred.
- Long service lives. Printers stay in use long after their vendor stops shipping firmware. Every flaw found after that point is permanent.
- Flaws that firmware cannot fix. CVE-2024-51978 required a change to Brother's manufacturing process; devices already sold need a workaround. Only a password change protects them.
- Shared components. When a flaw is in a third-party library such as the Treck TCP/IP stack, every maker that used it must ship its own fix, and Treck could not publish its client list because of non-disclosure agreements, according to Ripple20 coverage. Users have no way to know their printer was affected unless the brand says so.
The fix is mostly process: an inventory of every device and its firmware, a patch schedule with a fast track for critical flaws, contracts that say who patches, and a retirement date for every model. The fleet checklist below sets it out.
How can you tell if a printer has been hacked?
Signs include unexplained print jobs, admin passwords that no longer work, changed scan destinations, LDAP or DNS settings, new user accounts, unexpected reboots, firmware versions you did not install, and network traffic to places a printer should never go. Most compromises are silent, so regular configuration checks and network monitoring matter more than symptoms.
For home users
- Strange pages printing on their own, or the printer waking at odd hours.
- The admin password no longer works, or settings have changed.
- The printer appears reachable from outside your home. Free online port checkers will tell you whether your public address answers on printer ports; your router's port forwarding page tells you why.
- A firmware version that does not match the vendor's latest, or one you do not recognise.
For businesses
- Configuration drift. Compare each device's configuration with a known-good baseline: admin accounts, scan destinations, LDAP and SMTP servers, DNS, enabled protocols and installed apps.
- Network behaviour. Printers connecting to the internet, scanning internal ranges, or authenticating to servers they never used before.
- Directory and file server logs. Printer service accounts logging in from machines that are not printers.
- Host-side indicators. New printer drivers or ports on Windows systems, printers added to Linux hosts without anyone requesting them, and shell processes spawned by print filters.
- Firmware integrity. On devices with signed firmware and integrity checking, alerts about failed verification or recovery events.
What to do if you think a printer has been compromised
- Disconnect it from the network and the phone line.
- Preserve evidence if it is a business incident: export logs and configuration before you reset.
- Rotate every credential stored on the device, starting with directory, email and file share accounts.
- Reinstall the latest firmware from the vendor's official source, then reset to factory defaults.
- Reconfigure from scratch with a new unique admin password and only the services you need.
- Review the rest of the network, starting with what the printer could reach.
How do you secure a home printer?
Change the admin password, update the firmware and turn on automatic updates, keep the printer off the internet, turn off features you do not use, put it on a guest or IoT network if your router supports one, and wipe it before you sell or recycle it. That takes under an hour and blocks most attacks.
Home printers are mostly at risk from three things: a default password, old firmware, and accidental internet exposure. The steps below cover all three.
Figure 9. Printer hardening checklists for home, office and fleet.
The home checklist
- Change the admin password on the printer's web page to a long, unique one. Do it even if the printer came with a password on a label.
- Update the firmware now through the printer's menu or the vendor's app, and turn on automatic updates.
- Keep it off the internet. Do not set up port forwarding for the printer. Turn off UPnP on your router unless you need it. Use the vendor's cloud print feature if you need to print remotely.
- Turn off what you do not use. Fax, Wi-Fi Direct (the printer's own hotspot), and any web services or cloud features you never use.
- Use a guest or IoT network for the printer if your router supports it, so a compromised printer cannot reach your laptops. Make sure your computers can still print to it.
- Secure the vendor account with a strong password and multi-factor authentication if you use cloud or mobile printing.
- Keep your computers patched, including Windows updates, which fix the print spooler, and Linux and Unix package updates, which fix CUPS.
- Wipe the printer before selling or recycling it. Use the factory reset and any data erase option in the menu.
- Replace printers that no longer get firmware updates when you next need a new one.
Your router and Wi-Fi protect the printer too; the router guide explains how to lock them down.
How do you secure printers in an office or a managed fleet?
Inventory every device and its firmware, set unique admin passwords, put printers on their own network segment that only print servers can reach, print through a print server, use least-privilege accounts for scanning, turn off unused services including fax, encrypt and wipe storage, patch on a schedule, send logs to your SIEM, and write patch duties into vendor contracts.
Figure 10. Defend printers in layers; each layer catches what the one before misses.
The office checklist
- Inventory every printer and MFP: model, serial number, firmware version, location, network address, storage present, fax line present, owner, and end-of-support date.
- Set a unique admin password per device, stored in your password manager, and use named admin accounts with directory integration where supported.
- Segment printers onto a printer VLAN. Allow print traffic only from print servers, admin traffic only from a management network, and outbound traffic only to the services each device needs.
- Print through print servers rather than direct from laptops. It centralises driver control, blocks cross-site printing from user browsers, and makes logging easier.
- Use least-privilege accounts for scan to email, scan to folder and LDAP, with access only to the specific folders and lookups they need.
- Turn off unused services: fax and its phone line, FTP, Telnet, SNMP versions 1 and 2c (use SNMPv3), WSD, SLP and other discovery protocols you do not use, USB printing, and unused web services.
- Turn on HTTPS and IPPS and replace self-signed certificates where your device supports it.
- Encrypt storage, turn on automatic image overwrite, keep job retention short, and use pull printing for sensitive work.
- Patch the hosts: apply the print spooler and CUPS defences described above on every computer and server.
- Log and monitor: printer syslog to your SIEM, alerts on configuration changes and unusual connections.
- Plan disposal: wipe or destroy storage before any device leaves, with a certificate per serial number.
The fleet manager checklist
Fleet managers and managed print providers run hundreds or thousands of devices across sites and brands. The extra controls are about scale and accountability.
- Firmware baseline per model, with a target version and a date it must be reached.
- Fast track for critical and exploited flaws, measured in days, separate from routine updates.
- Central configuration policy pushed from the vendor's fleet management tools, with alerts when any device drifts from it.
- End-of-support tracking for every model, and a replacement plan before the date.
- Contracts that state who patches, how fast, and who wipes devices at the end of a lease, and that require vendors to publish advisories with CVE numbers and, increasingly, SBOMs (software bills of materials, explained below).
- Incident playbooks for a compromised printer, including credential rotation for every account stored on it.
- Procurement requirements for new devices: signed firmware, secure boot, encrypted storage, unique default credentials, a published support period, and a vulnerability disclosure program.
What is in the printer firmware supply chain?
Printer firmware is assembled from many suppliers: chip and board vendors, embedded operating systems, licensed network stacks, font and page interpreters, and open source libraries, integrated by the brand or an ODM, then signed and distributed. A flaw anywhere in that chain ends up in the printer, and a fix has to travel back down it.
An ODM, or original design manufacturer, designs and builds products that another company sells under its own brand. A software supply chain is everything that goes into a piece of software: your own code, third-party and open source components, build tools and the path from build to user. For printers, that chain is long.
Figure 11. Every stage of the firmware supply chain adds code, and every stage can delay a fix.
Shared components spread one flaw across many brands
The Ripple20 vulnerabilities are the clearest printer example. Treck's TCP/IP library had spread through direct and indirect licensing for about two decades; the 19 flaws affected vendors including HP, and the fix required Treck's stack version 6.0.1.67 or later to be built into each maker's firmware. Because Treck's customer list was covered by non-disclosure agreements, the only way a maker could answer "are we affected?" was to know what was inside its own firmware.
Printers also share image decoders, font engines, TLS libraries, compression libraries and web server components. The Faxploit bugs were in image parsing; the Printing Shellz bug was in font parsing; the Pwn2Own entries above are concentrated in parsers and network services. Many of those components come from somewhere else.
Firmware integrity is a supply chain control
The 2011 Columbia research showed what happens when a printer accepts firmware without checking a signature: anyone who can reach the update path can replace the device's software. Signed firmware, verified at boot and at update time, closes that path, and HP's 2015 changes to its enterprise LaserJets, which check that firmware is digitally signed by HP before it loads, are an example of the control done right. The signing keys then become the crown jewels: they need hardware protection, separation of duties and monitoring.
HP's own survey found that 51 percent of IT decision makers could not confirm whether printers had been tampered with in the factory or in transit. Customers are starting to ask makers to prove what is inside the device and that it has not been altered.
SBOMs make the chain visible
A software bill of materials (SBOM) is a machine-readable list of every component in a piece of software, with versions and suppliers. For a printer maker, an SBOM per firmware build answers two questions the morning a new component flaw is announced: which of our models contain it, and which customers need a fix. For fleet managers, a vendor SBOM turns "is our printer affected?" from a support ticket into a lookup. Our guide to SBOMs for firmware and IoT devices covers formats and practice, and CycloneDX vs SPDX compares the two main standards.
What should printer makers do?
Ship unique, random default credentials, sign and verify firmware, harden the parsers that read untrusted input, turn off risky services by default, generate an SBOM for every build, fix vulnerable components fast, publish a support period and advisories with CVE numbers, and make updates automatic.
The manufacturer checklist
- Unique, truly random default credentials per device, never derived from serial numbers, MAC addresses or other readable data, and a forced change at setup. The Brother case shows the cost of getting this wrong: a manufacturing change and millions of devices that firmware cannot fully fix.
- Signed firmware with secure boot, keys held in hardware, and rollback protection so old vulnerable firmware cannot be reinstalled.
- Harden every parser. Fuzz fonts, PDF, PostScript, PCL, PJL, image and fax decoders and every network service continuously. Run static analysis (SAST) on all first-party code. Where possible, move parsers to memory-safe languages or sandbox them. Fifteen of the 17 Pwn2Own entries we counted were memory-safety bugs.
- Secure defaults. Admin interface over HTTPS only; raw port 9100, FTP, Telnet, SNMPv1 and v2c, and unused discovery protocols off unless the customer turns them on; no services listening on the WAN side of any built-in network features.
- Least privilege inside the device. Run network services and interpreters with the minimum rights they need, so one bug does not hand over the whole printer.
- Protect stored credentials. Never send stored passwords to a changed server without re-entering them, which defeats pass-back, and encrypt credentials and storage at rest.
- An SBOM for every firmware build, driver and app, kept current and shareable with customers on request.
- Track and fix third-party components, prioritised by whether the vulnerable code is actually reachable in your firmware, and do not wait months for a supplier's next SDK release when a patched build of the component exists.
- Automatic, staged firmware updates on by default, with clear release notes.
- A published support period at the point of sale, honoured, with clear end-of-life notices.
- A vulnerability disclosure program, with advisories that list CVE numbers, affected models and fixed versions.
- Secure the build and signing infrastructure: protected build servers, reproducible builds where possible, and provenance records for every release.
How Safeguard helps printer and MFP makers
Safeguard gives printer makers, ODMs and managed print providers one platform for the software supply chain inside their devices: SBOMs for firmware, drivers and apps, binary and image upload scanning, reachability-ranked vulnerabilities, SCA, SAST and secrets scanning on code they own, fix pull requests gated by their own CI, Gold patched packages, supplier risk and compliance evidence.
Printer security teams usually stitch this together from separate tools for composition analysis, static analysis, secrets, binary analysis, SBOM management and audits, each with its own inventory. Safeguard covers it end to end, from the first inventory to the merged fix and the audit evidence, on one inventory and one vulnerability model.
Figure 12. Where Safeguard fits in a printer maker's pipeline.
See inside the binaries and images you ship and receive
In the Safeguard console, Integrations > Upload App Binary accepts compiled software and images: Windows EXE, MSI and DLL files, Android APK and iOS IPA apps, VM and disk images in QCOW2, VMDK, VHD and OVA formats, and container tarballs. Safeguard unpacks the file, identifies embedded libraries and versions, maps them to known vulnerabilities with EPSS (a score estimating the probability a flaw will be exploited) and known exploited enrichment, records licences, and produces a project with a full SBOM.
For a printer business that covers the Windows driver installer and DLLs, the mobile print apps, a fleet management server shipped as a VM image, and the containerised cloud services behind remote printing. Printer firmware images usually need to be extracted into their files first with your normal firmware tooling; Safeguard does not claim to parse every raw firmware format, and heavy obfuscation, encryption or unusual compression can give incomplete results. More in our guide to scanning app binaries with Safeguard and the explainer on binary composition analysis.
Source scanning where you own the code
The embedded web server, scan and fax services, cloud agent, drivers, apps and build system are code you own, and that is where SSRF, command injection and authentication bypass bugs are born. Safeguard runs software composition analysis (SCA) across the full transitive dependency tree, SAST on your own code, and secrets scanning that catches signing keys, cloud tokens and hard-coded credentials before they ship in an image. Vulnerabilities are ranked by reachability, EPSS and known exploitation, so the team fixes the flaw in an internet-reachable web service before a theoretical issue in a module that is never loaded.
SBOMs your customers and fleet managers can use
Every project produces an SBOM you can export in CycloneDX or SPDX, so you can answer a fleet manager's "what is in this firmware?" for every release, and "are we affected?" the morning a new flaw lands in a shared TCP/IP stack, font engine or image library. SLSA, Sigstore and in-toto attestation add build provenance, which speaks directly to the tamper question HP's survey raised.
Fix it without waiting for upstream
This is where Safeguard differs most from a scanner. For code in your repositories, Griffin, Safeguard's remediation model, opens a fix pull request that your own CI builds and tests before anyone merges it, and your policies decide what can merge. For open source components where upstream has not shipped a fix, or the fix sits in a version your platform cannot move to yet, Safeguard Gold provides patched, hardened builds of the packages it carries, and Safeguard can fork and patch the vulnerable package so your next firmware or driver build pulls a fixed version. For a printer team waiting on a supplier's next SDK release, that is the difference between fixing a parser flaw in the next build and leaving a fleet exposed for a quarter.
Supplier risk and compliance evidence
Safeguard's Third-Party Risk Manager lets you request SBOMs from chip vendors, ODMs, OS and stack suppliers and licensed engine vendors, score them and monitor them continuously (more on Safeguard TPRM). The compliance suite collects evidence for frameworks such as ISO 27001 and SOC 2 from the same scans. Guard adds runtime guardrails on the Linux servers that build, sign and distribute firmware, inspecting system calls, library calls and network egress, so a tampered build or update server is noticed early.
Deploy where your code lives
Firmware source and signing infrastructure are sensitive. Safeguard runs as SaaS in US and EU regions, in your own VPC, or fully on premises, and can run in Zero AI mode with no language model in the loop, or with your own model. Safeguard is SOC 2 Type I and Type II audited.
How this compares
Device makers often pair a source SCA tool such as Snyk or Black Duck with a firmware analysis specialist such as Finite State, which lists firmware analysis and SBOM among its product security offerings, or ONEKEY, which offers product cybersecurity and compliance for connected devices, plus a GRC platform for audits. Those are capable products, and for deep analysis of raw firmware images the specialists go further. Safeguard's case is consolidation and remediation: one inventory across source, binaries and images, fix pull requests and patched packages instead of tickets, and supplier risk and compliance evidence from the same data. For wider context, read our firmware supply chain security guide and hardware supply chain trust boundaries.
This post is part of a device security series. The laptop teardown follows the firmware inside a laptop, the post on fabless chip designers covers how the chips themselves are designed and supplied, and IoT device security fundamentals covers the wider class of connected devices printers belong to.
The takeaway
Most printers are not hacked by brilliant attackers. They are hacked through admin passwords nobody changed, print ports nobody meant to expose, firmware nobody updated, and service accounts with far more access than scanning needs. Owners can close most of that gap quickly: a unique admin password, current firmware, no internet exposure, unused services off, a separate network segment, least-privilege scan accounts, and a wipe before disposal.
The harder part belongs to the people who build printers. A printer is a stack of parsers and network services, much of it from other suppliers, written in languages that make memory-safety bugs easy, and it stays in service for years. Makers who know what is in every firmware build, fix their own code and shared components without waiting, sign what they ship, and support devices for as long as people use them, are the ones whose printers stop appearing in research papers and contest results.
Try it yourself
Click through these interactive Safeguard demos to see the steps described in this guide.
Test yourself
Which printer attacks matter most, and why?
The cheap ones: default or weak admin passwords, known firmware bugs left unpatched, printers exposed to the internet, and stored credentials that leak to anyone with admin access. Parser zero-days and fax exploits are real but need more skill, so fix the cheap ones first.
What did Pwn2Own show about printer firmware?
Researchers took over Canon, HP and Lexmark printers at every event reviewed, in 2022, 2024 and 2025. Fifteen of the 17 entries with a named bug class were memory-safety bugs, so printer firmware is still largely memory-unsafe code parsing network input.
What was PrintNightmare?
CVE-2021-34527 was a 2021 remote code execution flaw in the Windows Print Spooler. It let an authenticated user run code as SYSTEM, including on domain controllers, and the fix needed both patches and configuration changes because Point and Print could be left insecure.
How did the 2024 CUPS vulnerabilities chain together?
CVE-2024-47176 let a remote attacker make cups-browsed add a fake printer, and CVE-2024-47177 in cups-filters let the attacker's commands run. Exploitation needs user interaction, because someone must print to the fake printer.
How did Faxploit reach a network through a printer?
Check Point showed in 2018 that a specially crafted fax could take over an HP OfficeJet, using flaws CVE-2018-5924 and CVE-2018-5925. It travelled over the phone line, which bypasses every firewall, email filter and web proxy, and needed only the fax number.
What can an attacker get from an office printer besides paper?
Queued and retained print jobs, scans, fax logs, an address book, and saved passwords for the email server, file shares and company directory. Someone who gets the drive from a disposed device can also read old documents.
Why do printer fleets stay unpatched?
In HP's 2025 survey only 36 percent of IT teams promptly applied printer firmware updates. Mixed fleets, manual or outsourced updates, fear of breakage, long service lives and shared components all compound it, and some flaws, like Brother's default password issue, cannot be fixed in firmware at all.
Frequently asked questions
Can a printer be hacked?
Yes. Network printers are computers with operating systems, web servers and network services. Researchers at Ruhr University Bochum found all 20 models they tested vulnerable to at least one attack, and Canon, HP and Lexmark printers have been exploited at Pwn2Own. The common causes are default passwords, unpatched firmware and printers exposed to the internet.
Can someone hack my printer from the internet?
Only if it is reachable from the internet, directly or through a port forward, or if someone on your network opens a malicious document or web page that sends data to it. Studies have found hundreds of thousands of exposed printers. Check your router has no port forwards to the printer, turn off UPnP, and keep firmware updated.
How do I know if my printer has been hacked?
Look for print jobs you did not send, an admin password that no longer works, changed settings such as scan destinations or DNS, unexpected reboots, and firmware you did not install. In an office, compare configurations with a baseline and watch for printers connecting to unusual addresses. Many compromises are silent, so prevention matters more than symptoms.
Is PrintNightmare still a risk?
It is fixed on Windows systems that have received updates since July 2021 and that keep Point and Print restricted, as Microsoft advises. Unpatched or misconfigured systems remain at risk, and the spooler remains privileged code. Disable it where nobody prints, especially on domain controllers, and consider Windows protected print mode.
Do printers store copies of what they print and scan?
Many office multifunction printers store jobs, scans, fax logs and address books on internal storage, and some retain jobs for reprinting. In 2010 CBS News recovered medical records and police records from used copiers. Turn on storage encryption and overwrite, keep retention short, and wipe or destroy storage before disposal.
Should I turn off fax on my multifunction printer?
If you do not use fax, yes: unplug the phone line and disable the feature. In 2018 Check Point's Faxploit research showed that an HP OfficeJet could be taken over by sending it a crafted fax, bypassing firewalls entirely. If you need fax, patch fax devices first and segment them, or use a central fax service.
How often should I update printer firmware?
Turn on automatic updates where available, and check monthly otherwise. Apply critical or actively exploited fixes within days. HP's survey found only 36 percent of IT teams promptly apply printer firmware updates, which is why known flaws stay exploitable. Replace printers that no longer receive updates.
Is it safe to print over Wi-Fi at home?
Yes, if your Wi-Fi uses WPA2 or WPA3 with a strong passphrase, the printer has a unique admin password and current firmware, and it is not exposed to the internet. Turn off Wi-Fi Direct if you do not use it, and put the printer on a guest or IoT network if your router supports one and your computers can still reach it.
Sources
- Rapid7, Multiple Brother devices: multiple vulnerabilities (fixed)
- SecurityWeek, New vulnerabilities expose millions of Brother printers to hacking
- Rapid7, Xerox VersaLink C7025 multifunction printer pass-back attack vulnerabilities (fixed)
- CSO Online, New HP MFP vulnerabilities show why you should update and isolate printers
- SecurityWeek, Canon patches 7 critical vulnerabilities in small office printers
- BleepingComputer, Lexmark warns of RCE bug affecting 100 printer models, PoC released
- Zero Day Initiative, Pwn2Own Toronto 2022 day one results
- Zero Day Initiative, Pwn2Own Ireland 2024 day one results and day three results
- Zero Day Initiative, Pwn2Own Ireland 2025 day one results and day two results
- Rapid7, CVE-2021-34527 (PrintNightmare) analysis
- Yarden Shafir and Alex Ionescu, PrintDemon (CVE-2020-1048)
- Microsoft Learn, Windows protected print mode
- Microsoft Learn, End of servicing plan for third-party printer drivers on Windows
- Simone Margaritelli, Attacking UNIX systems via CUPS, Part I
- Elastic Security Labs, Cups Overflow: when your printer spills more than ink
- Akamai, When CUPS runneth over: the threat of DDoS
- Check Point, Faxploit
- The Hacker News, What maximum a remote attacker can do just by having your fax machine number
- Microsoft Security Response Center, Corporate IoT: a path to intrusion
- Threatpost, YouTuber PewDiePie promoted via 50K hacked printers
- Security Affairs, Experts hacked 28,000 unsecured printers to raise awareness of printer security issues
- Müller, Mladenov, Somorovsky and Schwenk, SoK: Exploiting Network Printers, IEEE Symposium on Security and Privacy 2017
- Computerworld, HP LaserJet printers vulnerable to attacks, researchers warn
- CIO, HP adds protection against firmware attacks to enterprise printers
- BankInfoSecurity, Millions of connected devices have exploitable TCP/IP flaws
- Legaltech News, report on the CBS News copier investigation
- TruthOrFiction, digital copiers retain images of documents
- Quocirca, Print Security Landscape 2025 press release
- SecurityBrief UK, Poor printer security leaves firms exposed, according to HP report
- ONEKEY, home page
- Finite State, about