The Low Down

The Low Down

By Low LevelTechnology
Download on the App Store

All The Low Down episodes

  • Java4Shell, NGINX2Root & the Week Everything Went Down

    Recorded Wednesday, November 18, 2026.

    Java and NGINX had a bad week. So did anyone trying to catch a flight.

    The Low Down Podcast is presented by Maze.

    https://mazehq.com/lowdown

    X: https://twitter.com/Maze_Security

    Ed and Matt get into Java4Shell, NGINX2Root, and why a week of outages had people asking about a cyberattack. Then we talk about what happened when Optimizely’s A/B testing scripts started serving malware.

    Today on The Low Down

    Java4Shell and “vibe uncoding”

    Java is everywhere. Ed gets into which apps teams need to worry about, and Matt brings up Calif’s pitch for using AI to find old code we can rip out.

    Cloudflare’s guidance for teams patching Java

    https://blog.cloudflare.com/java4shell-waf-protection

    NGINX2Root and what the numbers actually count

    Five research teams and two HTTP/2 bugs. Matt pulls apart the big exposure numbers and gets to the question a version scan can’t answer for you. Is HTTP/2 actually turned on?

    Wiz’s research and disclosure timeline

    https://www.wiz.io/blog/nginx-http2-rce-vulnerabilities

    NGINX security releases

    https://nginx.org/2026.html

    Why did so many services go down?

    An FAA ground stop, bank disruptions, and a reported NSA network outage all landed in the headlines. We go through what was known about the outages and why old systems are so hard to keep running.

    BBC’s outage roundup

    https://www.bbc.com/news/articles/c6z0kly2lll7o

    WIRED on how to tell an outage from a cyberattack

    https://www.wired.com/story/how-to-tell-if-its-a-cyberattack-or-just-an-outage/

    Optimizely and the scripts running on your site

    A tool for testing button colors and headlines ended up serving malicious JavaScript on customer sites. We compare what Scott Helme saw with the timeline Optimizely published and ask who’s keeping tabs on the scripts we load.

    Report URI investigation

    https://blog.report-uri.com/worlds-largest-brands-infected-with-malware-via-optimizely/

    Optimizely’s incident report (updated since recording)

    https://pages.optimizely.com/rs/947-RKE-048/images/Security%20Incident%20Report%20-%20Optimizely%20Web%20Experimentation%20-%20Nov%2017%202026.pdf

    Thanksgiving wrap

    We’re taking the week off for Thanksgiving. Thanks for hanging out with us these first six months.

    The Low Down is presented by Maze.

    LinkedIn: https://www.linkedin.com/company/mazehq/

    X: https://twitter.com/Maze_Security

    Follow The Low Down

    https://www.instagram.com/lowdown.pod

    Ed / Low Level

    https://www.instagram.com/lowleveltv/

    https://www.linkedin.com/in/lowleveltv/

    https://www.youtube.com/@LowLevelTV

    Matt Johansen / Vulnerable U

    https://www.instagram.com/mattjayy/

    https://www.linkedin.com/in/matthewjohansen/

    https://www.youtube.com/@VulnerableU

    58 min
  • This SSH Bug Accepted Wrong Passwords, AI Exploits, Tiny Keys & a Surprise Guest

    Recorded on Friday, November 13, 2026

    The Low Down Podcast is presented by Maze.
    https://mazehq.com/lowdown
    X: https://twitter.com/Maze_Security

    257 A's should not get you into an SSH account.

    John Hammond joins Ed and Matt to talk about N-able, ScreenConnect, and a rough week for remote support. Then we get into the SSH login bug, OpenAI's Linux kernel research, FortiClient, and eMudhra's tiny private keys. There's also some World of Warcraft, because apparently we needed a break.

    Today we're talking about:

    John Hammond, N-able, and ScreenConnect
    John walks through the RMM mess, including SCKit and the difference between a vulnerable server extension and a vulnerable technician client. We talk about how remote support can spread an infection, then get into incident-response burnout.
    N-able report: https://www.huntress.com/blog/n-able-disables-cloud-services-following-disclosure-of-able2shell-exploit-chains
    SCKit research: https://www.huntress.com/blog/sckit-worm-rogue-screenconnect
    Client advisory: https://www.connectwise.com/company/trust/security-bulletins/2026-09-08-screenconnect-bulletin

    Exploitation in the wild and local AI
    Why doesn't every scary bug turn into an internet-wide fire? We talk about exploit conditions, GreyNoise's PHP observations, a separate AI-assisted intrusion report, and running local models for security research.
    GreyNoise PHP tracking: https://php.events.greynoise.io/
    Operation Snowball: https://malbearlabs.com/posts/operation-snowball-inside-an-autonomous-ai-intrusion-campaign

    The SSH login bug
    Ed explains how OpenSSH and musl can turn a password longer than 256 bytes into a successful login on affected configurations. We follow the asterisks, the old fixes, and the practical problem of figuring out which devices are exposed.
    Original write-up and comic: https://sshhole.com/

    OpenAI's kernel bugs and Daybreak
    OpenAI reports 948 working local privilege escalation exploits. We unpack what that number means, the API-credit cost, and the research pipeline. Then we ask what happens to AI security vendors when the model companies start offering the tools themselves.
    Linux kernel research: https://openai.com/index/a-thousand-fewer-paths-to-root/
    Daybreak Red Team: https://openai.com/daybreak/red-team/
    Daybreak Defense Factory: https://openai.com/daybreak/defense-factory/

    FortiClient and the security-vendor problem
    A FortiClient/FortiPAM exploit demo leads into a bigger argument about edge devices, old toolchains, and why preventable bugs keep shipping in security products.
    Researcher's demo: https://www.linkedin.com/posts/kevin-joensen_cve-2026-121529-or-fortinet2shell-as-we-ugcPost-7526030522238697472-3JP7/

    eMudhra's weak private keys
    One private key was 4. We talk about the HSM migration mistake behind eMudhra's weak intermediate-certificate keys, what the hardware was supposed to protect, and why checking the result matters.
    Company incident report: https://bugzilla.mozilla.org/show_bug.cgi?id=2083449#c82

    World of Warcraft and logging off
    Alliance or Horde, busy-dad gaming, and finding time to play something without turning it into another job.

    Which is harder to keep up with right now: the vulnerabilities, the patches, or the AI product launches?

    Linked reports may include later updates.

    ----------

    0:00 Intro and John Hammond
    3:59 N-able and ScreenConnect attacks
    13:48 Exploitation in the wild
    17:28 Jailbroken AI and local models
    22:48 The 257-character SSH login bug
    32:42 OpenAI’s Linux kernel exploit research
    44:03 Daybreak and AI security vendors
    49:55 FortiClient and edge security
    58:14 eMudhra and the private key of 4
    1:04:13 World of Warcraft

    ----------

    Listen on Spotify!
    https://go.lowdownpod.com/spotify

    The Low Down is presented by Maze.
    LinkedIn: https://www.linkedin.com/company/mazehq/
    X: https://twitter.com/Maze_Security

    Follow Us!
    https://www.instagram.com/lowdown.pod

    Ed / Low Level
    Instagram: https://www.instagram.com/lowleveltv/
    LinkedIn: https://www.linkedin.com/in/lowleveltv/
    YouTube: https://www.youtube.com/@LowLevelTV

    Matt Johansen / Vulnerable U
    Instagram: https://www.instagram.com/mattjayy/
    LinkedIn: https://www.linkedin.com/in/matthewjohansen/
    YouTube: https://www.youtube.com/@VulnerableU

    John Hammond
    YouTube: https://www.youtube.com/@_JohnHammond

    1 hr 11 min
  • Binance's Fake Interview Hack, Railway's AI Incident, Hacked Freezers & a Critical Linux Bug

    The Low Down Podcast is presented by Maze.
    https://mazehq.com/lowdown
    X: https://twitter.com/Maze_Security

    The "signup test" was probing for Kubernetes secrets.

    Ed and Matt get into Railway's AI testing incident, hacked refrigeration systems, and fake job interviews targeting developers. Then Ed works through a Linux kernel bug before we finish with ARTEX and a warning about a prebuilt Docker image.

    Today we're talking about:

    Railway and the AI testing incident
    We follow the logs, the apologies, and the personal credit card that helped identify the employee. Josh Manders says every probing attempt was blocked and nothing was accessed; Railway later described it as an employee's personal project. Ed and Matt have plenty to say about authorization and the "my agent did it" explanation.
    Employee's account (https://x.com/jakerunzer/status/2117474337682780643)
    Josh's account (https://x.com/joshmanders/status/2117530391183507484)
    Railway's statement (https://x.com/JustJake/status/2117746864745578719)
    Credit-card exchange (https://x.com/vxunderground/status/2117656394396778956)

    Hacked freezers, blood banks, and exposed controls
    More refrigeration attacks lead into the practical problems of securing operational systems. We talk about slow patch approvals, unclear ownership, and what getting a management interface off the public internet actually involves.
    Supermarket report (https://www.bbc.com/news/articles/cq1kd1nee80eo)
    Blood-storage report (https://www.theguardian.com/technology/2026/nov/04/france-national-blood-service-hack-cyber-attack-europe)

    Binance and fake job interviews
    Binance's account of an employee compromised through a fake interview leads into how fake recruiters target developers. Matt walks through separate malicious coding-test and VS Code examples, including the pressure to run unfamiliar code while an interviewer watches. We discuss keeping those exercises off your everyday machine.
    Binance report (https://www.reuters.com/world/binance-says-hackers-used-fake-job-interview-compromise-computer-before-63-2026-11-03/)
    MetaPlay case study (https://github.com/alexhraber/metaplay-fake-interview-supply-chain-incident)
    VS Code case study (https://ransom-isac-landing.pages.dev/blog/contagious-interview-vscode-to-rat/)

    Linux TCP, MPTCP, and exploit math
    Ed explains a kernel use-after-free, the demonstrated MPTCP exploit, and why reliable exploitation is more complicated than the severity score suggests. We get into the conditions an attacker needs, patching, and some increasingly elaborate math.
    Go Security Team advisory (https://groups.google.com/g/golang-announce/c/PCoQcdxoScw/m/aD-MgXEvAAAJ)

    ARTEX and the Docker image warning
    CrowdStrike reports on ARTEX in real intrusions. Separately, researchers raise backdoor concerns about a specific prebuilt Docker image. We look at what they found and Jason Haddix's SpikeStrip experiment for tripping up AI pentesting tools.
    CrowdStrike research (https://www.crowdstrike.com/en-us/blog/vapor-panda-uses-artex-to-scale-global-campaign/)
    Image warning (https://x.com/GrahamHelton3/status/2117782531856581004)
    SpikeStrip announcement (https://x.com/Jhaddix/status/2117342447445627141)

    If you work with grocery-store or other operational networks, we'd like to hear what getting these systems off the internet looks like on the ground.

    Linked reports may include later updates.

    —————-

    Listen on Spotify!
    https://go.lowdownpod.com/spotify

    The Low Down is presented by Maze.
    LinkedIn: https://www.linkedin.com/company/mazehq/
    X: https://twitter.com/Maze_Security

    Follow Us!
    https://www.instagram.com/lowdown.pod

    Ed / Low Level
    Instagram: https://www.instagram.com/lowleveltv/
    LinkedIn: https://www.linkedin.com/in/lowleveltv/
    YouTube: https://www.youtube.com/@LowLevelTV

    Matt Johansen / Vulnerable U
    Instagram: https://www.instagram.com/mattjayy/
    LinkedIn: https://www.linkedin.com/in/matthewjohansen/
    YouTube: https://www.youtube.com/@VulnerableU

    1 hr 7 min
  • Black October Recap, OpenCode’s Stolen API Keys, AI Distillation & the LiteLLM Supply-Chain Attack

    Welcome to The Low Down, the best show on the internet for hackers

    Black October has real bugs, stolen keys, and a lot of very loud headlines.

    Ed and Matt sort through PHP2Shell, AI security claims, and the difference between a scary vulnerability and one attackers can actually use. Harry Wetherald, co-founder and CEO of Maze, joins us to talk about finding the bugs that matter.

    ——————————


    Today we're talking about:
    PHP2Shell and the Black October panic
    We get into working exploits, questionable proof-of-concepts, patching, and the reporting around attacks. Ed breaks down why configuration and exploit reliability matter, while Matt revisits a report he shared that was later corrected.

    • Original research: https://calif.io/php2shell/
    • Exploit validation: https://www.vulncheck.com/blog/php2shell-vulnerabilities
    • Unit 42 report and updates: https://unit42.paloaltonetworks.com/cve-2026-109618-cve-
    • 026-109619/
    • DREAM intrusion research: https://dreamgroup.com/blog/chained-php2shell-exploit-roots-25-servers

    AI security with Harry from Maze
    Harry joins us to talk about what AI can do for security teams, how to decide whether a vulnerability is exploitable in your environment, and why producing more findings isn't enough.

    • Maze: https://mazehq.com/

    Brotli, Node.js, and GoAhead
    A decompression bug puts Brotli in the headlines, then GoAhead gives us a lesson in broken TLS. We talk through the technical details and the conditions that turn these bugs into real problems.

    • Node.js security release: https://nodejs.org/en/blog/vulnerability/october-2026-security-releases
    • GoAhead research: https://foxio.io/blog/choose-your-backend-three-tls-bugs-in-goahead

    OpenCode, stolen keys, and distillation
    OpenCode's Console incident leads to a discussion about stolen API keys and model distillation. Matt revisits his initial response and the distinction between stealing keys and what attackers intended to do with them afterward.

    • OpenCode announcement: https://x.com/opencode/status/2111519637049319818
    • Matt's clarification: https://x.com/mattjay/status/2111554937452990593

    LiteLLM, Railway, and the update problem
    More supply-chain trouble raises an uncomfortable question: how quickly should you update when new releases can carry the attack? We get into credential exposure, dependency hygiene, and the tradeoffs of waiting.

    • LiteLLM attack-chain report: https://docs.litellm.ai/blog/security-incident-oct-21-2026-attack-chain
    • Railway incident report: https://blog.railway.com/p/preliminary-report-october-security-incident

    WPS, pre-boot bugs, and AI bank access
    We cover Broadcom WPS research, an EDK II PXE vulnerability, and AI agents accessing bank accounts. The banking discussion turns on whose permission was given: the customer's or the bank's.

    • WPSplinter research: https://qkaiser.github.io/broadcom-wps/
    • EDK II advisory: https://github.com/tianocore/edk2/security/advisories/GHSA-99cj-7hg9-mp4m
    • AI banking report: https://www.axios.com/2026/10/22/ai-agents-banks-unauthorized-accounts-access

    We finish with the noise around security news, incident-response fatigue, and what happens when every week is treated like the end of the internet.

    Recorded October 28, 2026.

    Linked reports may include later updates.

    Listen on Spotify!
    https://go.lowdownpod.com/spotify

    The Low Down is presented by Maze.
    LinkedIn: https://www.linkedin.com/company/mazehq/
    X: https://twitter.com/Maze_Security

    Follow Us!
    https://www.instagram.com/lowdown.pod

    ——————————

    Ed / Low Level
    Instagram: https://www.instagram.com/lowleveltv/
    LinkedIn: https://www.linkedin.com/in/lowleveltv/
    YouTube: https://www.youtube.com/@LowLevelTV

    Matt Johansen / Vulnerable U
    Instagram: https://www.instagram.com/mattjayy/
    LinkedIn: https://www.linkedin.com/in/matthewjohansen/
    YouTube: https://www.youtube.com/@VulnerableU

    1 hr 25 min
  • Cybersecurity Experts REACT to the php chaos... (and MORE)

    Welcome to The Low Down, the best show on the internet for hackers

    Listen on Spotify!
    https://go.lowdownpod.com/spotify

    The Low Down is presented by Maze.
    LinkedIn: https://www.linkedin.com/company/mazehq/
    X: https://twitter.com/Maze_Security

    ----------------

    Today we're talking about:

    PHP2Shell and the Disclosure Fight
    Ed and Matt are at OneCon in Las Vegas as PHP2Shell sets off a chaotic week. We unpack the research, shaky AI exploit claims, Chainguard's early fixes and whether disclosure timelines can keep up. Dan Lorenc and the PHP Foundation give different accounts of the rollout.
    https://calif.io/php2shell/
    https://www.linkedin.com/posts/danlorenc_i-wanted-to-share-what-happened-with-the-share-7518756991520120832-QX2_/
    https://thephp.foundation/blog/2026/10/21/todays-php-security-release-and-how-were-strengthening-php-security/
    Kyle Polley's embargo discussion: https://x.com/kpolley/status/2113069186318577666

    Your Certificate Authority's Private Key Was 4
    One of eMudhra's private keys was literally four. We talk tiny keys, certificate trust, the company's exit from SSL/TLS and what happens when clients still trust compromised certificates.
    https://scotthelme.co.uk/eight-bits-of-trust-emudhras-ca-keys-cracked-in-milliseconds/
    https://emudhra.we-have-a-problem.com/
    https://emudhra.com/en/press-release/emudhra-to-exit-global-ssl-tls-certificate-business
    https://www.mitmproxy.org/posts/releases/mitmproxy-12.3/

    Linux Bugs and the Power of a Pattern
    DirtyFrag, research credit and the latest kernel discussion lead into why Ed thinks AI is useful for spotting use-after-free patterns.
    DirtyFrag research: https://github.com/V4bel/dirtyfrag
    KSMBD background: https://docs.kernel.org/filesystems/smb/ksmbd.html

    Does AI Favor Attackers or Defenders?
    SentinelLABS' Tom Hegel makes the case for keeping powerful tools accessible to defenders. SentinelOne's Steve Stone explains why finding more bugs isn't the whole story: attack paths, speed, scale and the physics of an intrusion still matter.

    Meet the Threat Hunting World Champion
    Jason Walker joins us straight from his OneCon win. We get into his move from graphic design to security, CTF strategy, AI-assisted detection engineering and why the fundamentals still matter.
    OneCon: https://www.onecon.io/

    ----------------

    Follow Us!
    https://www.instagram.com/lowdown.pod

    Ed / Low Level
    Instagram: https://www.instagram.com/lowleveltv/
    LinkedIn: https://www.linkedin.com/in/lowleveltv/
    YouTube: https://www.youtube.com/@LowLevelTV

    Matt Johansen / Vulnerable U
    Instagram: https://www.instagram.com/mattjayy/
    LinkedIn: https://www.linkedin.com/in/matthewjohansen/
    YouTube: https://www.youtube.com/@VulnerableU

    Tom Hegel / SentinelLABS
    LinkedIn: https://www.linkedin.com/in/tom-hegel/

    Steve Stone / SentinelOne
    LinkedIn: https://www.linkedin.com/in/stevenstone618/

    Jason Walker / tresscross
    LinkedIn: https://www.linkedin.com/in/jasonwalker777/
    Blog: https://tresscross.blog/

    1 hr 28 min
  • Cybersecurity Experts BREAK DOWN the shocking OpenAI Spying Incident, Crypto Backdoors, AND MORE

    Welcome to The Low Down, the best show on the internet for hackers

    Listen on Spotify!
    https://go.lowdownpod.com/spotify

    The Low Down is presented by Maze.
    LinkedIn: https://www.linkedin.com/company/mazehq/
    X: https://twitter.com/Maze_Security

    Follow Us!
    https://www.instagram.com/lowdown.pod

    Heading to OneCon 2026? SentinelOne’s cybersecurity conference runs October 20–22 at the ARIA Resort & Casino in Las Vegas. Use code oc26lowdown for a discount on registration.
    https://www.onecon.io/

    Today we're talking about:

    AI Got Better While We Were Arguing About It
    Matt returns from SecTor, Ed explains his monitor-firmware experiment, and we discuss what AI can do for security research—and the human work it still takes.
    https://www.youtube.com/watch?v=wIe3eDfGKUo
    https://blog.cloudflare.com/cyber-frontier-models/
    https://hacks.mozilla.org/2026/05/behind-the-scenes-hardening-firefox/

    Your Hardware Wallet Has Extra Hardware
    Kim Zetter's reporting reveals seed-stealing Ledger implants, including devices from authorized retailers, and separate Trezor tampering from unauthorized sellers. Who planted them remains unknown.
    https://www.zetter-zeroday.com/who-is-planting-backdoors-in-ledger-crypto-wallets/
    https://www.zetter-zeroday.com/trezor-wallets-also-compromised-with-implants/

    Someone Hacked the Supermarket Fridges
    French supermarkets lose food after a cyberattack. We discuss exposed refrigeration controllers, a date-derived login code, and the separate Ahold Delhaize outage.
    Reuters report: https://srnnews.com/french-supermarket-chain-intermarche-hit-by-cyberattack-150-stories-targeted/
    https://claroty.com/team82/research/freeze-the-controller-defrost-the-food-uncovering-vulnerabilities-in-danfoss-refrigeration-controllers
    https://www.pentestpartners.com/security-blog/you-dont-need-a-nation-state-to-warm-up-a-fridge/
    https://www.bleepingcomputer.com/news/security/retail-giant-ahold-delhaize-takes-systems-offline-after-cyberattack/

    If AI Can Clone Your App, What Are You Selling?
    ArtCraft and Reverse Engineer Anything spark a debate about paid software, reliability and support. Fireship's FilmCraft test puts the hype under pressure.
    https://www.kdnuggets.com/the-rise-of-ai-software-clones-why-rea-and-artcraft-are-going-viral
    https://github.com/morluto/rea
    https://www.youtube.com/watch?v=wajoP_5Cpw8

    The People You Call After Ransomware
    We discuss Edward Dubrovsky's reported arrest and the separate MonsterCloud case. Prosecutors allege MonsterCloud secretly paid attackers while promising special decryption tools. These are allegations, not convictions.
    https://krebsonsecurity.com/2026/10/fbi-arrests-founder-of-ransomware-negotiation-firm/
    https://www.bleepingcomputer.com/news/security/ransomware-recovery-ceo-charged-over-secret-ransom-payments/
    https://www.justice.gov/opa/pr/known-cybersecurity-expert-and-owner-florida-ransomware-remediation-company-charged

    OpenAI Finds a Recorder Behind the TV
    An employee is accused of planting a recorder in an executive conference room. We unpack the alleged chatbot trail, a second device and the problem of insider access.
    https://www.bbc.com/news/articles/cwg5jq4vlddzo
    https://x.com/OpenAINewsroom/status/2110095403198849132

    A Jailbreak That Isn't Quite a Jailbreak
    hrtowii's iOS research involves Apple's Security Research Device signing system. The author explains the limits and says the bug has been patched.
    https://www.hrtowii.dev/blog/making-a-not-jailbreak-in-3-weeks

    1 hr 17 min
  • Oops, all vulns! (Citrix NetScaler problems, the FBI ShinyHunters saga cont’d and MORE)

    Welcome to The Low Down, the best show on the internet for hackers

    Listen on Spotify!
    https://go.lowdownpod.com/spotify

    The Low Down is presented by Maze.
    LinkedIn: https://www.linkedin.com/company/mazehq/
    X: https://twitter.com/Maze_Security

    Follow Us!
    https://www.instagram.com/lowdown.pod

    Heading to OneCon 2026? SentinelOne’s cybersecurity conference runs October 20–22 at the ARIA Resort & Casino in Las Vegas. Use code oc26lowdown for a discount on registration.
    https://www.onecon.io/

    00:00 Coming up
    00:33 Intro and Matt’s SecTor keynote
    02:52 Citrix NetScaler zero-days and disclosure
    09:25 Breaking down the NetScaler exploits
    19:04 Sponsor: Maze
    20:41 Citrix’s response and responsible disclosure
    25:06 Why edge devices are so hard to defend
    30:54 MikroTik RouterOS vulnerabilities
    35:33 Cisco SD-WAN authentication bypass
    38:37 Zimbra command injection
    42:22 ShinyHunters arrest and conflicting claims
    50:42 Reacting to the FBI’s warning
    55:05 Open-weight AI, Anthropic and GLM-5.3
    1:07:35 Wrap-up and OneCon

    Today we're talking about:

    Your NetScaler Warning Came From Reddit
    Finding out about NetScaler zero-days through Reddit is not the enterprise support experience Matt had in mind. Ed walks through watchTowr's command injection and memory overflow writeups, including missing protections in the binary they tested. Then we get into Google's hunting guidance and why a patch alone won't kick an attacker out.
    https://www.reddit.com/r/Citrix/comments/1wqjk9a/netscaler_leak/
    https://labs.watchtowr.com/oh-look-the-foot-gun-went-off-again-citrix-netscaler-preauth-command-injection-cve-2026-88771/
    https://labs.watchtowr.com/here-we-go-again-citrix-netscaler-dtls-preauth-memory-overflow-cve-2026-88772/
    https://cloud.google.com/blog/topics/threat-intelligence/defending-against-active-exploitation-of-citrix-netscaler-adc-and-gateway-appliances

    MikroTik and the User Named -2
    MikroTik joins the pile with an HTTP underflow and a separate SSH authentication chain. The username "-2" showing up in attack logs gets our attention. Ed explains how assumptions between different parts of a system can become an exploit. Matt has been saying "don't roll your own crypto" for his entire career. Apparently we still need the reminder.
    https://github.com/cisagov/CSAF/blob/develop/csaf_files/OT/white/2026/icsa-26-272-06.json
    https://cert.pl/en/posts/2026/09/vulnerabilities-in-mikrotik-routeros-actively-exploited/

    Cisco's Admin API Lets the Wrong People In
    A specially encoded request can give an unauthenticated attacker admin access to Cisco's SD-WAN Manager API. We talk about what that means for a corporate network, and the difference between connections that need to be public and management interfaces that don't. Cisco patched its managed cloud service. If you run an affected instance yourself, the upgrade is still your job.
    https://www.cisco.com/c/en/us/support/docs/csa/cisco-sa-sdwan-webauth-xr8beuuU.html

    Zimbra: An Email Goes In, a Shell Command Comes Out
    A crafted email can trigger command execution through Zimbra's SNMP notification system on affected configurations, without anyone clicking anything. We trace the route from incoming mail to a shell command and look at Microsoft's account of the attacks. Ed has a fairly blunt suggestion for developers: stop handing user input to Bash.
    https://www.microsoft.com/en-us/security/blog/2026/09/30/unauthenticated-command-injection-on-internet-facing-mail-servers-tracking-cve-2026-73570/

    ShinyHunters Meets the FBI's Camera
    Dutch police arrest an alleged ShinyHunters leader, and the FBI looks straight into the camera with a message for whoever's left. We untangle the arrest timeline, conflicting gang statements and the Pokémon branding. Matt brings us back to the PeopleSoft campaign and why following a group's methods is useful even when its PR is impossible to trust.
    https://krebsonsecurity.com/2026/09/dutch-police-arrest-reformed-hacker-in-shiny-hunters-investigation/
    https://www.fbi.gov/video-repository/shinyhunters-arrested-092926.mp4/view
    https://www.securityweek.com/shinyhunters-defiant-after-fbi-calls-on-members-to-come-forward/
    https://cloud.google.com/blog/topics/threat-intelligence/shinyhunters-renewed-mass-exploitation-campaign-targeting-oracle-peoplesoft

    If the Attacker Gets AI, So Do I
    Anthropic's GLM-5.3 report sets off a disagreement. If attackers get powerful, freely available models, Matt wants to know why defenders don't get the same advantage. Ed thinks the way our systems are built still favors the attacker. We get into reports of safeguards getting in the way of incident response, the incentives behind calls for regulation, and who we'd trust to write the rules.
    https://www.anthropic.com/research/glm-5-3-and-the-spread-of-advanced-cyber-capabilities
    https://huggingface.co/blog/security-incident-july-2026

    1 hr 9 min
  • ShinyHunters’ FBI Breach, OpenAI’s Bug Bounty Drama, and the Rise of ClickFix

    Welcome to The Low Down, the best show on the internet for hackers

    Listen on Spotify!
    https://go.lowdownpod.com/spotify

    The Low Down is presented by Maze.
    LinkedIn: https://www.linkedin.com/company/mazehq/
    X: https://twitter.com/Maze_Security

    Follow Us!
    https://www.instagram.com/lowdown.pod

    Today we're talking about:

    ShinyHunters Has a Reputation to Protect
    ShinyHunters says it hacked the FBI to protect its business. Apparently, even an extortion crew needs good PR. We get into why reputation matters to people who make their money threatening companies, and Ed takes a look at the PeopleSoft bugs that may explain the way in. Some of the exposed records reportedly belong to the FBI's own hacking unit.
    https://www.rapid7.com/blog/post/etr-active-exploitation-of-oracle-peoplesoft-zero-day-cve-2026-35273/
    https://www.404media.co/fbi-hack-exposed-fbis-own-hacking-unit-remote-operations-shinyhunters/
    https://www.theregister.com/cyber-crime/2026/09/25/shinyhunters-tells-the-reg-we-hacked-the-fbi-to-protect-our-business/5299250

    OpenAI's Very Unwelcome Pull Request
    How far do you go to prove a bug? Hacktron used an employee's Codex account to open a PR in OpenAI's internal repo. Ed thinks that demonstrates the access they already had. Matt sees why the security team wants to know what else they touched. The $6,500 payout got everyone talking. S1r1u5_ says the headache was a disclosure process that sent them to lawyers and eventually a journalist.
    https://www.hacktron.ai/blog/hacking-openai
    https://x.com/S1r1u5_/status/2101296199261429813
    https://x.com/LiveOverflow/status/2101252692635004997
    https://x.com/TalBeerySec/status/2101560506242761097
    https://x.com/alexstamos/status/2101747329699402198

    So What Do You Run This Stuff In?
    Two awkward headlines for anyone running other people's code: containers are getting easier to escape, and KVM has bugs too. Matt wants to know what he's supposed to do with that. Ed explains why sharing a kernel matters, and why moving to a smaller attack surface still buys you something. We also end up worrying about who will understand the machinery underneath all the AI-written code.
    https://depthfirst.com/research/containers-are-no-longer-safe

    About That AI Hacking Story
    Ed has had enough of the headlines. Trying a few web attacks and pulling off the Hugging Face exploit are very different things. Matt's bothered by agents on ordinary research tasks deciding to try breaking into websites. We look at what Transluce actually found: the public records don't show those attempts succeeding. Both have questions about who was watching.
    https://transluce.org/agent-activity

    "But I Would Never Fall for ClickFix"
    Four in five ClickFix payloads intercepted by Push in 2026 came through search engines. You go looking for an answer and get instructions for infecting your own computer. Matt has a pile of data on how the lures are changing. Then there's the placeholder domain in old documentation that someone registered and turned into a fake verification page. But we would never fall for any of this, obviously.
    https://pushsecurity.com/blog/the-state-of-clickfix-by-detection-data
    https://www.manifold.security/blog/third-party-com-placeholder-clickfix

    1 hr 31 min
  • Who’s Reading Your ChatGPT Chats? Flock Camera Secrets and HBO’s Reddit Account Spreads Malware

    Welcome to The Low Down, the best show on the internet for hackers

    Listen on Spotify!
    https://go.lowdownpod.com/spotify

    The Low Down is presented by Maze.
    LinkedIn: https://www.linkedin.com/company/mazehq/
    X: https://twitter.com/Maze_Security

    Follow Us!
    https://www.instagram.com/lowdown.pod

    Today we're talking about:

    The ChatGPT Group Chat
    "I thought me and Sam were homies." Project Lily has us talking about the humans reading ChatGPT conversations, Moxie Marlinspike's take on chat interfaces, and who else is in the room when you're telling a chatbot your problems.
    https://www.404media.co/inside-project-lily-the-humans-reading-your-chatgpt-chats/
    https://confer.to/blog/2025/12/confessions-to-a-data-lake/

    Flock, Meet the Hardware Hackers
    Someone pulled down a Flock camera and copied its storage. Ed wants to separate the physical-access findings from the remote-hacking claims. Also, "Flock bad" is apparently a content strategy now.
    https://www.404media.co/hackers-stole-flocks-camera-software-revealing-how-the-company-tracks-cars-and-people-2/

    Pixel's Very Unhelpful CVE
    A modem vulnerability, signs of targeted exploitation, and very few details. Cue a conversation about basebands, spyware, and the price of a good phone exploit.
    https://www.bleepingcomputer.com/news/security/google-fixes-actively-exploited-android-zero-day-on-pixel-devices/
    https://source.android.com/docs/security/bulletin/pixel/2026/2026-09-01

    Revolut
    "A bank suffered a data breach by not getting hacked." Fake official requests got customer data handed over. Matt gets into the market for stolen government accounts and why a familiar email address isn't enough.
    https://www.securityweek.com/revolut-data-breach-5-months-680-high-profile-accounts-3m-ransom/

    HBO's Reddit Account
    The account was real. Someone else was running the ads. A compromised HBO business account on Reddit becomes a delivery system for a ClickFix malware campaign.
    https://www.hudsonrock.com/blog/hbo-max-ads-on-a-compromised-reddit-account-exposed-a-massive-pasteswitch-clickfix-operation

    Cisco Got Owned by an Email
    An email security gateway. An email-triggered SQL injection. "The call is coming from inside the house."
    https://sec.cloudapps.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-esa-inj-2bLVGmhX

    Your MRI Results Are Malware
    Iranian operators using fake medical results and personalized messages to deliver CHOSEN BRICK spyware. This one gets uncomfortably specific.
    https://www.ncsc.gov.uk/news/iranian-cyber-targeting-of-dissidents-activists-and-journalists

    Passkey-Themed Phishing
    That word "themed" is doing a lot of work. We walk through the device-code trick that gets victims to authorize an attacker's access.
    https://www.microsoft.com/en-us/security/blog/2026/09/09/passkey-themed-social-engineering-leads-identity-cloud-compromise/

    Inflatable PLCs
    CISA wants critical infrastructure operators thinking about cyber decoys. We're into canary tokens. A full deception operation for an already understaffed team takes a little more convincing. Ed has a business idea.
    https://www.cisa.gov/sites/default/files/2026-09/using-cyber-decoys-to-strengthen-detection-and-response_508c.pdf

    1 hr 18 min
  • LG Caught Spying in Your Living Room, Miami Crypto King Gets Life, and the AI Apocalypse Debate

    Welcome to The Low Down, the best show on the internet for hackers

    The Low Down is presented by Maze.

    LinkedIn: https://www.linkedin.com/company/mazehq/

    X: https://twitter.com/Maze_Security Follow Us!

    https://www.instagram.com/lowdown.pod

    This week we're diving into the surveillance nightmare coming from your living room, catastrophic phishing campaigns targeting crypto users, and Microsoft's record breaking patch cycle that defenders need to know about.

    Today we're talking about:

    LG's Privacy Nightmare: Your TV is Spying on Everything Gamers Nexus drops a two hour expose revealing LG TVs are actively recording and transcribing all room audio even when listening mode is disabled and the device is powered off. Breaking down how they got root shell access and found plain text transcriptions of private conversations being cached on the device with four gigs of memory ready to phone home. Why the microphone continues recording even when unplugged from the network and what ACR automatic content recognition really means for your privacy.

    The Economics of Smart TV Surveillance LG executives caught on stage at advertising conferences bragging about owning the glass and extending the advertising experience across all devices on your home network. How they're doing network scans to enumerate every Windows PC, phone, and IoT device in your household and correlating that with what you're watching. Why the profit margins on TV advertising are in the 60s with SaaS level multiples while TV hardware prices have bottomed out over the last 20 years.

    The Remote Code Execution Problem Nobody's Talking About Beyond the privacy concerns, LG TVs have multiple unpatched remote code execution vulnerabilities that turn your living room microphone into a potential threat actor playground. Why teachers are panicking about 85 of these devices deployed in classrooms and the student privacy implications. How Chinese threat actors have proven they'll exploit residential devices for proxy networks and credential theft.

    Trezor's Catastrophic Phishing Campaign Trezor customers hit with one of the nastiest phishing campaigns of the year where the call was literally coming from inside the house. Breaking down how their official email provider was breached with valid TLS certificates and domain signing making the phishing email appear completely legitimate. Why this was perfectly timed after the cold card STM 32 entropy vulnerability had everyone primed to believe their hardware wallets were compromised.

    1 hr 17 min