Robert Floyd @_RobertFloyd

Tweets

799
Followers

2K
Following

586
Likes

643

Robert Floyd @_RobertFloyd

3 months ago

A massive data breach exposed ~16 billion credentials from major platforms like Google, Apple, and Facebook. This isn’t old data—it’s fresh, weaponizable, and dangerous. ➡️ If your org isn’t ready for passkeys and enforcing MFA, you're already behind. Risk is never static.…

0 0 1 123 0

Robert Floyd @_RobertFloyd

4 months ago

Vulnerabilities aren’t just patches—they’re lessons. As a CISO, I push for root-cause analysis and CVE transparency to eliminate entire classes of flaws. How do you manage vulns? #SecureByDesign #VulnerabilityManagement"

0 0 1 44 0

Robert Floyd @_RobertFloyd

4 months ago

Built-In > Bolt-On If your product roadmap needs a security feature added later, you’re already behind. “Secure by Design” isn’t just good practice — it’s the future of procurement. Buyers are watching. So are regulators. #AppSec #ProductSecurity #SBOM #CyberResilience

0 0 1 27 0

Robert Floyd @_RobertFloyd

4 months ago

Cyber Debt Is Real Technical debt is expensive. Cyber debt is catastrophic. Poor design choices compound over time, exposing users and infrastructure to avoidable risk. You either pay upfront in secure design… Or you pay later in headlines. #SecureDesign #Cybersecurity #TechDebt

0 0 1 19 0

Robert Floyd @_RobertFloyd

4 months ago

Your AI system is only as secure as its data supply chain. Poisoned datasets, expired domains, third-party model risk — attackers don’t need to break your model if they can compromise your inputs. Secure the pipeline. Verify the provenance. #AIsecurity #DataIntegrity…

0 0 1 21 0

Robert Floyd @_RobertFloyd

4 months ago

Using a foundation model trained on unknown data is like buying food with no ingredient list. If the vendor can’t attest to: ✅ Clean training data ✅ Source certifications ✅ Poisoning defenses ...you’re ingesting risk. #AItrust #CISOs #SupplyChainSecurity

0 0 1 27 0

Robert Floyd @_RobertFloyd

5 months ago

Everyone's talking about AI risk. But no one agrees on what it actually means. MIT's AI Risk Repository analyzed 65 taxonomies and found 1,612 risk categories. We don’t just have an AI alignment problem — we have a vocabulary crisis. #AIsecurity #AIGovernance #RiskTaxonomy

0 0 1 29 0

Robert Floyd @_RobertFloyd

5 months ago

AI is already used in: • Surveillance • Disinformation • Exploit development • Deepfake fraud • Social engineering And these threats don’t require model breakthroughs — just API access. The future isn’t just offensive. It’s automated. #AdversarialAI #GenAI #CyberThreatIntel…

0 0 1 35 0

Robert Floyd @_RobertFloyd

6 months ago

AI is transforming cybersecurity — and threat actors know it. Here are 5 AI threats every CISO should have on their radar: Prompt Injection Malicious inputs that hijack LLMs and bypass controls. Model Poisoning Compromising training data to insert backdoors. Shadow AI…

0 0 3 179 1

Robert Floyd @_RobertFloyd

6 months ago

AI models don’t just make predictions. They create attack surfaces. From evasion to prompt injection, the adversarial landscape for GenAI is expanding fast. If your AI governance doesn’t include AML threat modeling, you’re not managing risk — you’re guessing. #AIsecurity…

0 0 2 65 0

Robert Floyd @_RobertFloyd

6 months ago

AI models don’t hallucinate risk — they hide it. As we embed LLMs into workflows, security leaders must shift from protecting data at rest to protecting decisions in flight. #AIsecurity #LLM #CyberRisk

1 0 2 89 0

Robert Floyd @_RobertFloyd

6 months ago

Nation-state threats aren’t new. But our detection strategies can’t stay old. What matters now isn’t attribution — it’s containment velocity and layered trust boundaries. #CyberResilience #ThreatDetection #ZeroTrust

0 0 1 28 0

Robert Floyd @_RobertFloyd

6 months ago

AI in cybersecurity isn’t just about defense — it’s about decision velocity. The real opportunity: Augmenting your team’s ability to triage, correlate, and respond at scale. It won’t replace your analysts. But it will force you to rethink what they’re best at. #AI…