*make sure to check your spam folder for the signup acceptance email.
|
One of the challenges of being productive is always making sure you get enough sleep each night. It's been hard to keep my energy up with renovations going on at home, and so much noise around me while I'm trying to work. That's definitely been my challenge over the last couple of weeks, and it will be for the next five weeks of renovations too. So here's to looking forward to that. Headline: The Co-Pilot Who Turned on His Captain
This story is still developing. We don't know the motive. We only have some of the facts so far. But whatever the motive, if the reports are accurate, someone in a trusted position, inside a secured environment, broke that trust to attack a colleague. That's really concerning in a safety-critical place like the cockpit of a plane. Many lives are at stake, not only on the plane, but on the ground too if it had crashed into a populated area. Source: flydubai statement on FZ1073 (Gulf News) Source: Co-pilot stabbed captain, tried to crash flight, Netanyahu says (CBS News) Source: Initial Saudi probe (Al Jazeera) Insider Scoop: Everywhere at OnceRicheline Fung, CISA
Evester Edd, Peace Corps
Both of these fall under what the subreddit calls overemployed. I wrote an insider threat advisory on this about three years ago: employees holding multiple jobs, particularly when they're remote and can juggle them. There are two sides. The employee often sees nothing wrong with it. But there are conflict of interest issues, and intellectual property can be shared between the companies, accidentally or on purpose, because one person is working all those jobs. It's a moral grey area if the contract doesn't define it. With government jobs it's very explicit. You need prior approval to work anything beyond your main job. And billing overlapping hours on contracts just doesn't make sense. That's ethically wrong. Source: Former Homeland Security employee pleads guilty to $250,000 fraud scheme (USAO D.C.) Source: Former federal employee sentenced for time and attendance scheme (USAO D. Md.) Source: Former federal employee charged with wire fraud, theft and false statements (USAO D. Md.) Crime RoundupThe CEO who emptied her own credit union.
Source: Former JAFCU head pleads guilty to three federal counts (WLBT) Source: Jackson Area Federal Credit Union closes; Five Star Credit Union assumes members and shares (NCUA) The border supervisor and the Newegg trade-ins.
Source: Border Patrol supervisor in Maine accused of modifying government computers (Portland Press Herald) The ex-soldier who paid serving soldiers.
The head of IT and the $8 million in phones.
Source: IBAC charges NGV employee with theft and misconduct in public office as part of Operation Sonder (IBAC) Source: IBAC charges senior National Gallery of Victoria employee over alleged $8 million phone scheme (ABC News) These again are all examples of people using insider access to do something illegal and harm the organisations they're part of. The US Army one is really interesting, though. Duan was paying serving soldiers to get information for him, using them as his aggregators. If the allegations hold, the soldiers taking his money are almost as guilty as he is. In a military setting there's no reason to accept payment to go and get information. It would be a different story if he'd coerced them, or socially engineered them into just helping him. AI Insider Threat Watch: OpenAI Shows Three Safety Researchers the Door
I'm not trusting this story until there's some outside validation. OpenAI did its own investigation and then let these people go. So what were its internal methods for handling the safety researchers' work? Did they raise these concerns internally? Do they have whistleblower protection, and was it completely ignored? Did the researchers think the information was critical enough to share outside? And what exactly was shared that warranted letting them go? This doesn't look good for OpenAI, or for the track record frontier labs have with handling safety researchers. Source: OpenAI parts ways with researchers who allegedly shared confidential information (Wall Street Journal) Source: OpenAI cuts ties with three safety researchers, WSJ reports (TechCrunch) Source: OpenAI ignored employees who warned it wasn't doing enough about security (New York Times) Also this week: the agents that made their own public repos
Again, another risk exposed by AI agents doing something unexpected, and undirected by the people running them. With arguably over-permissive access, we're going to see more and more data leaks like this through the rest of this year and into next. Source: How AI agents exposed developer screenshots from leading tech companies (Glow Labs) Source: AI coding agents leaked 13,000 internal company screenshots to public GitHub repos (Help Net Security) Mitigation Corner: Nine Resignations in One Day
This is a textbook case of people potentially taking information to a new employer, where it will benefit them. How the new employer reacts is really critical. And catching it is only possible when you've got monitoring and visibility in place to track leavers. Especially sudden leavers. When someone resigns, especially suddenly, go back through their history and look at what they did before they handed it in. Source: Citizens Bank sues rival lender, alleges trade-secret theft in poaching of CT mortgage staff (Hartford Business Journal) Source: Citizens Bank accuses SoFi of poaching 30 mortgage employees (Banking Dive) Trusted Profession Watch: The Fraud Investigator in the Dolomites
It's just surprising when people think they can get away with this kind of thing. And it makes it worse for the rest of us who rely on insurance and compensation schemes. This is still alleged, and there's more to come out. But taken at face value, it looks pretty bad. TriviaA new study used Stasi archives to estimate what East Germany gained from economic and industrial espionage against the West. By how much did it boost East Germany's economic output in the late 1980s? A) 0.5% B) 2.1% C) 7.4% D) 18% (Answer at the bottom.) Corporate News: The FBI Finally Ticks the Box
It's interesting that they're saying they've now reached full compliance with the federal minimum standard. Were they outside their time requirement to be fully compliant? Does this play into the recent breach? Was there any insider activity there? Minimum compliance is always just the first tick in the box. What else are we doing in this space to improve insider threat visibility? Source: FBI partners with Everfox to deploy advanced insider threat protection across FBI (FBI) Source: How CNSSD 504 defines user activity monitoring (National Counterintelligence and Security Center) UpdatesQuick follow-ups on stories from earlier issues. ShinyHunters and the FBI. In Issue #24 we covered ShinyHunters' claim to hold the FBI's staff data. The FBI has since confirmed the breach (see Corporate News). Dutch police also arrested a 24-year-old in Amsterdam on 15 September, a week before the FBI claim, as a suspect in their ShinyHunters investigation. Security researchers say he was convicted in 2023 for hacking and extortion and was on supervised release. When he was arrested, he was working as an offensive security lead at a security firm. Source: Dutch police arrest "reformed" hacker in ShinyHunters investigation (KrebsOnSecurity) Source: Dutch police arrest convicted hacker in ShinyHunters investigation (SecurityWeek) Rippling v Deel. In Issue #19 we covered Rippling's claim that Deel paid a Rippling employee, Keith O'Brien, to spy from inside its Dublin office. On 25 September a federal judge in California refused Deel's bid to strike O'Brien's testimony, and refused to revisit his ruling letting Rippling's racketeering and trade secret claims proceed. Deel's own counterclaims were sent to private arbitration. Source: Deel dealt new setbacks in Rippling case over alleged spying (Bloomberg) What's OnBefore the Insider Acts: Understanding the Critical Pathway to Insider Risk, an Everfox webinar.
Insider Risk: From Reactive to Proactive, an INSA webinar.
Insider Risk Community of Practice Networking Drinks, Melbourne, hosted by the Australian Cyber Collaboration Centre.
On the horizon: the Adelaide edition of the networking drinks is on Monday 7 December at the Crowne Plaza. ResourcesA few places worth your time if you want to go deeper on insider risk.
Worth bookmarking:
And my own: Insider Threat Awareness Training on YouTube, a free course with certification. Trivia AnswerC, about 7.4%. In manufacturing alone, value added rose by 22.3%, based on 1988 data, roughly 4.1 billion euros in 2020 prices. Source: Espionage boosted East Germany's economic output by 7.4% (Phys.org) |
*make sure to check your spam folder for the signup acceptance email.