SecTor 2025 | DriveThru Hacking: Now with Delivery
The video presents a research project titled “Drive‑Thru Hacking: Now with Delivery,” demonstrating how dash‑cam devices can be compromised when a vehicle stops at a drive‑through. The team, led by Benjamin So and colleagues, scanned over 1,000 Wi‑Fi SSIDs in Singapore, purchased more than 20 dash‑cams from 16 brands, and built an automated tool that discovers, authenticates, extracts, and exfiltrates video footage within the brief window a car spends in a drive‑through lane. Their findings show that eight‑in‑ten dash‑cams are installed, many manufactured in Asia, and a majority share identical hardware and firmware. Fifteen of the twenty‑two tested units shipped with the same default password, and four brands used immutable passwords, making unauthenticated Wi‑Fi access trivial. The researchers also uncovered unprotected API ports, unauthenticated upload endpoints, and hard‑coded credentials that enable full root control, video download, and even remote battery sabotage. Notable demonstrations include spoofing a trusted device’s MAC address to bypass pairing, using port‑knocking on custom API ports (7778/7779) to retrieve video and audio streams, and uploading a CGI web shell to gain root on model K. They also highlighted a misconfiguration where dash‑cam mobile apps expose live feeds to anyone, revealing owners’ routes, home addresses, and private conversations. The work underscores a systemic lack of security hygiene in the dash‑cam market, where cost and video quality are prioritized over authentication and firmware integrity. Manufacturers face urgent pressure to enforce unique passwords, secure OTA updates, and disable default Wi‑Fi access, while consumers should treat dash‑cams as potential attack vectors that can compromise personal privacy and vehicle safety.
SecTor 2025 | Sharing Is Caring About an RCE Attack Chain on Quick Share
At SecTor 2025, SafeReach researchers Orya and Cohen unveiled a sophisticated remote‑code‑execution (RCE) attack chain targeting Google’s Quick Share, now available on Windows. The talk detailed how the team reverse‑engineered the Nearby Connections protocol, built a custom "QuickSniff" logger, and leveraged...
SecTor 2025 | Leading Across the Generations
The talk at SecTor 2025 focused on leading across generations in cybersecurity, emphasizing that technical expertise alone isn’t enough; leaders must grasp the cultural and motivational traits of each age cohort. The speaker highlighted how adoption timelines for technologies have shrunk—from...
SecTor 2025 | One Agent to Rule Them All: How One Malicious Agent Hijacks A2A System
The SecTor 2025 talk highlighted a growing security dilemma: multi‑agent generative‑AI systems, exemplified by Google’s A2A (Agent‑to‑Agent) protocol, can be weaponized by a single malicious agent that hijacks an entire automation ecosystem. The presenters, senior AI security researchers from Zenity...
Black Hat USA 2025 | ReVault! Compromised by Your Secure SoC
At Black Hat USA 2025, Cisco Talos senior researcher Firo Verity presented his findings on Dell’s Control Vault secure‑on‑chip (SoC) and how it can be compromised. Verity showed that the Broadcom‑based BCM58202 chip, used in over 100 Dell Latitude and Precision models,...
Black Hat USA 2025 | LLM-Driven Reasoning for Automated Vulnerability Discovery Behind Hall-of-Fame
The Black Hat USA 2025 talk introduced “Whisper,” a large‑language‑model‑driven system that automatically discovers vulnerabilities in stripped ARM64 binaries. The presenter, a researcher guiding an undergraduate team, explained how the tool earned a Hall of Fame award at Samsung...
Black Hat USA 2025 | Leveraging Jamf for Red Teaming in Enterprise Environments
The Black Hat USA 2025 session highlighted how adversary emulation teams can weaponize Jamf Pro—Apple’s enterprise‑device management platform—to conduct red‑team operations in Fortune‑500 environments. Speakers Lance Kane and Dan Mayer described Jamf’s prevalence in developer‑heavy organizations, its default “set‑and‑forget” configuration,...
Black Hat USA 2025 | 2 Cops 2 Broadcasting: TETRA End-To-End Under Scrutiny
Midnight Blue, a Dutch cyber‑security consultancy, presented at Black Hat USA 2025 a deep dive into the end‑to‑end encryption layer of the Tetra terrestrial trunked radio standard. Tetra, widely adopted for police, military and SCADA communications, has long kept its...
Black Hat USA 2025 | The 5G Titanic
The presentation likened the 5G architecture to the Titanic, arguing that, like the ship’s supposedly watertight compartments, 5G’s control‑plane and user‑plane are assumed to be isolated but in practice lack vertical sealing. The speaker outlined how the network’s design—AMF, SMF...
Black Hat USA 2025 | AI Agents for Offsec with Zero False Positives
Brendan Dolan‑Gavitt opened his Black Hat USA 2025 talk by warning that the promise of AI‑driven offensive security is haunted by a spectre of false positives. Drawing on his decade‑long experience in software security and recent work on GitHub Copilot,...
Black Hat USA 2025 | Protecting Small Organizations in the Era of AI Bots
The presentation at Black Hat USA 2025 focused on defending small, resource‑constrained organizations against the surge of AI‑driven bots. Citing the Impreva 2025 BadBot report, the speaker highlighted that 51% of all internet traffic is now non‑human, and that 80%...
Black Hat USA 2025 | Kernel-Enforced DNS Exfiltration Security
The presentation by independent researcher Vang Parnes focuses on the growing threat of DNS‑based command‑and‑control (C2) and tunneling techniques targeting Linux systems, especially in cloud environments. He outlines why DNS is the favored back‑door for advanced persistent threats (APTs), citing...
Black Hat USA 2025 | Burning, Trashing, Spacecraft Crashing
The Black Hat USA 2025 session, led by Mileno Star and Andre of Vision Space, highlighted the growing cyber‑risk landscape for space systems. With commercial constellations such as Starlink and OneWeb proliferating alongside renewed military satellite launches, the orbital environment now presents a...
Black Hat USA 2025 | Universal and Context-Independent Triggers for Precise Control of LLM Outputs
The Black Hat presentation introduced a novel class of prompt‑injection attacks called universal adversarial triggers, which allow attackers to hijack large language model (LLM) outputs with a single, reusable token sequence. By decoupling the malicious payload from the trigger,...
Black Hat Stories | Or Yair, Security Research Team Lead at SafeBreach
Ori Yair, security research team lead at SafeBreach, reflects on his Black Hat experiences, from his first nerve‑wracking speaking slot to his ongoing focus on Windows‑based vulnerability research. He frames the conference as a catalyst for turning technical curiosity into...
