Why cyber hygiene requires curious talent – Clea Ostendorf – ESW #355
Full Audio
View Show IndexSegments
1. Why cyber hygiene requires curious talent – Clea Ostendorf – ESW #355
Many years ago, I fielded a survey focused on the culture of cybersecurity. One of the questions asked what initially drew folks to cybersecurity as a career. The most common response was a deep sense of curiosity. Throughout my career, I noticed another major factor in folks that brought a lot of value to security teams: diversity.
Diversity of people, diversity of background, and diversity of experience. I've seen auto mechanics, biologists, and finance experts bring the most interesting insights and forehead-slapping observations to the table. I think part of the reason diversity is so necessary is that security itself is incredibly broad. It covers everything that technology, processes, and people touch. As such, cybersecurity workers need to have a similarly broad skillsets and background.
Today, we talk to someone that embodies both this non-typical cybersecurity background and sense of curiosity - Clea Ostendorf. We'll discuss:
- The importance for organizations to actively seek and welcome curious newcomers in the security field who may not conform to traditional cybersecurity norms.
- Strategies for organizations to foster an environment that encourages individuals with curiosity, motivation, and a willingness to challenge conventional norms, thereby promoting innovative thinking in addressing security risks.
Segment Resources:
Evolving Threats from Within - Insights from the 2024 Code42 Data Exposure Report
Announcements
Google has announced that they will be shutting down the Google Podcasts platform in mid-2024. To ensure that you don't lose access to the Security Weekly content you know and love, please make sure that you subscribe to your favorite podcasts feeds on an alternative platform such as Spotify, YouTube Music, Amazon Music, Apple Podcasts, Overcast, Podcast Addict, PocketCasts, or anywhere else you listen to podcasts! Visit securityweekly.com/subscribe to find the buttons to subscribe to each show now!
Guest
There are two types of people, those who follow the instructions while cooking, and those who read the ingredients and then go off feeling. Clea is the latter, which has served her well as she’s navigated various roles in technology and security from sales, to program management and most recently as Field CISO for Code42. Security is one part tech, but also needs to have compassion and partnership in order for any technology to be successful. Just as in cooking, Clea begins with the end in mind and brings a people and business focus to each engagement she is part of.
Hosts
2. Multi-Layered Defense Platforms and other terms we found in security press releases – ESW #355
This week, in the enterprise security news:
- Early stage funding is all the rage
- AI startups continue to pop out of stealth
- The buyer's market continues with more interesting acquisitions
- Purpose-built large language models for security
- Benchmarking LLMs for security
- GoFetch? More like... Get outta here (I couldn't think of anything clever)
- Crowdstrike and NVIDIA team up
- Why do people trust AI?
- What do Google Sheets and Carlos Sainz Jr. have in common?
All that and more, on this episode of Enterprise Security Weekly!
Announcements
Security Weekly listeners: Join the digital identity community at the ARIA Resort & Casino in Las Vegas, May 28 – 31. The 15th annual Identiverse will bring together over 3,000 security professionals for 4 days of world-class learning, engagement, and entertainment.
As a community member, receive 25% off your Identiverse 2024 tickets using code IDV24-SW25!
Register today: securityweekly.com/idv2024
Hosts
- 1. FUNDING: Finite State Raises $20 Million Growth Round to Secure Critical Infrastructure and Software Supply Chains
$20M growth round led by Energy Impact Partners. Generates SBOMs, does SCA, ingests data from other scanners, provides remediation guidance.
- 2. FUNDING: BlueFlag Security Emerges from Stealth with $11.5M Seed Funding to Address Critical Gaps in SDLC Security
$11.5M seed round led by Maverick Ventures and Ten Eleven Ventures. "BlueFlag Security offers a multi-layered defense platform that integrates identity security with open-source software risk management and developer tool posture management."
- 3. FUNDING: Software Supply Chain Security Leader Binarly Closes $10.5 Million Financing Led by Two Bear Capital
$10.5M round led by Two Bear Capital. Seems like an Eclypsium competitor. We had their CEO/founder on a while back, on episode 327.
- 4. FUNDING: Dymium Snags $7M to Build Data Security Platform with Secure AI Chat
$7 seed round led by Two Bear Capital. Another AI company trying to control input/output via a shim between users and LLMs.
- 5. FUNDING: Cybersecurity AI provider Auguria emerges from stealth to solve security operations data overload and cost problems
$6.5M seed round led by SYN Ventures and SentinelOne's S Ventures. "Auguria drives the transformation of traditional security operations by encoding generations worth of hard-won human security experience into artificial intelligence (AI) models capable of cutting through a sea of multi-vendor event data."
Virtual SOC analyst?
- 6. FUNDING: Tarsal Raises $6M in Seed Funding
$6M in seed funding led by Harpoon Ventures and Mango Capital. "One-click to build your security data lake."
- 7. FUNDING: Think Cyber Security raises £3m led by Fuel Ventures
£3m 'late' seed round, led by Fuel Ventures. Aiming to "deliver measurable secure behavior change" (security awareness training, then?) "Nudge your users to protect themselves..."
- 8. FUNDING: Leen announces $2.8M in pre-seed funding for its unified data API for security
$2.8M pre-seed round led by 11.2 Capital. My best guess at what they do is some abstracted meta-API of all your other security APIs?
- 9. FUNDING: LeakSignal clinches $1.6m in seed funding to enhance cyber in microservices
$1.6M seed round, led by Shasta Ventures. Another AI company trying to shim between users and LLMs.
- 10. ACQUISITIONS: GitLab Acquires Oxeye to Advance Application Security & Governance Capabilities
Oxeye will reportedly help GitLab accelerate its SAST roadmap, and augment its SCA and compliance tools.
- 11. ACQUISITIONS: Cyber Startup Wiz to Buy Gem Security For $350 Million
For the same price Zscaler picked up Avalor last week, Wiz gets a jump on the cloud detection and response space. Seems like this product is focused around useful, context-enhanced detections, and automated playbooks to do something about them.
- 12. ACQUISITIONS: JumpCloud Acquires Resmo
"JumpCloud Acquires Resmo for Next Gen IT, SaaS, and Cloud Asset Management"
I've been a JumpCloud fan for over a decade now. It was originally designed to be a security product, but pivoted into an SMB alternative to Microsoft. The fact that they're still around proves the market really needed something more versatile and easier to deploy and manage than Entra ID in smaller environments.
As JumpCloud expands with acquisitions, I wonder how far they can go, in building an alternative to Microsoft, Okta, and other massive complex platforms.
- 13. OPEN SOURCE: LLM4Decompile: Reverse Engineering: Decompiling Binary Code with Large Language Models
We're starting to see some very cool purpose-built LLM models!
- 14. VULNERABILITIES: GoFetch: Breaking Constant-Time Cryptographic Implementations Using Data Memory-Dependent Prefetchers
Another vuln with a marketing campaign. It's an interesting processor-focused sidechannel attack that amounts to privesc, much like Spectre. It's unlikely any attacker would ever need it or use it, but it has the potential to damage Apple M-series chip performance - the primary mitigation is to disable prefetch while executing cryptographic functions. M3 chips can disable prefetch, but M1 and M2 chips cannot.
For academic researchers, this is some cool research.
For defenders and IT folks, this is a nothing burger that might waste a lot of their time and hurt the performance of their users' systems.
- 15. AI NEWS: CSA’s Large Language Model (LLM) Threats Taxonomy is now ready for public peer review
The Cloud Security Alliance is working on an LLM threats taxonomy. Now is the time if you want to weigh in!
- 16. BENCHMARKS: Benchmarking the Security Capabilities of Large Language Models
This highlights the need for technology-based benchmarks for new technologies in security. We're already seeing purpose-trained LLMs, so it makes sense that we'd want to test existing LLMs for efficacy at various security-related tasks.
- 17. PARTNERSHIPS: CrowdStrike Collaborates with NVIDIA to Advance Cybersecurity with Generative AI
Normally, I'd roll my eyes at this as just an empty announcement we'll never see any tangible benefits from, like when Intel acquired McAfee. However, as we see purpose-built LLMs start to pop up, I'm intrigued about what Crowdstrike could potentially do with custom cybersecurity-focused LLMs!
- 18. ESSAYS: CISO Networks Decoded: What Works, What Doesn’t
- 19. PAPERS: What Motivates People to Trust ‘AI’ Systems?
- 20. SQUIRREL: Smooth scrolling in Google Sheets now available on desktop