VerticalSideways examines how statistical assumptions in automated classifiers can be exploited through multimodal steganography. Drawing on production-system research that led to a bug bounty, a fix, and a second bypass, the session connects image-based XSS, blind SSRF, graphical injection, and business-logic failures. It looks at what happens when automated systems and their human reviewers share the same blind spots.

For: AI security teams, bug bounty hunters, researchers, and students

A stolen session can give an attacker access without another password or MFA prompt. Dave Lewis traces the token-theft supply chain, from infostealer infections and log markets to access brokers, OAuth abuse, and adversary-in-the-middle campaigns. The session closes with a vendor-neutral checklist covering device verification, phishing-resistant authentication, session visibility and revocation, and credentials used by machines and AI agents.

For: Security practitioners and anyone responsible for access to an organization

Drone telemetry, improvised command links, and sensor data reveal more than flight paths. Using open-source observations and civilian research, Oksana Verbytska examines radio fingerprints, software supply chains, mapping overlays, and trust assumptions in autonomous systems. Recorded signals and analysis tools illustrate how researchers interpret, fingerprint, and simulate this data. The focus is analysis and defensive understanding, rather than exploitation techniques.

For: Drone hobbyists, radio enthusiasts, and government or military analysts

A small black window on a wall-mounted remote sparked a side project into how Japanese bidet controls communicate. Goku and Kalev explore infrared transmissions, reflections from everyday bathroom surfaces, and how devices respond to malformed signals. They share their experiments, unexpected findings, and practical lessons from investigating a familiar household device.

For: Anyone curious about hardware and everyday technology

A worldwide scavenger-hunt game, played through robots connected over 4G and 5G, became an unexpected research project. BusesCanFly follows the API flaws and stories uncovered along the way, then reflects on curiosity-driven hacking in a community shaped by financial incentives and LLMs. The session is an invitation to rediscover experimentation, persistence, and the pleasure of figuring things out.

For: A broadly technical audience, from curious newcomers to experienced hackers

Ports connect digital systems to physical supply chains. This talk examines vessel tracking, terminal operating systems, and satellite connectivity under the pressures of conflict and aging operational technology. Anastasia Kovalchuk considers how governments and private operators can model disruptions affecting shipping, grain exports, naval logistics, and emergency response—and what resilience means for the infrastructure behind them.

For: Government, military, maritime, and infrastructure practitioners

Space systems can contain the same software, network, and protocol flaws found on Earth. Ricardo Fradique uses demonstrations to examine weaknesses in mission control and onboard software, tracing the attack surface from user-facing systems to the RF links used for communication. The session challenges the assumption that distance, specialist hardware, or an orbital destination is itself a security control.

For: Security practitioners, hackers, and engineers curious about space systems

Espilon began as monolithic ESP32 firmware, where every new capability meant another rebuild and reflash. This session follows its redesign into a modular platform that dynamically loads signed ELF modules into IRAM and PSRAM. It covers binary relocation, fault isolation, and forensics, with examples across ESP32 C6, S3, and classic hardware. Wireless analysis, protocol fuzzing, and other modules show how the platform can support shared security research.

For: Embedded developers, security researchers, IoT testers, and educators

BobDaHacker presents the lifecycle of Daybreak Technologies, a honeypot operation designed to attract and document suspected North Korean IT-worker operatives. The investigation followed eleven people onboarded through job postings and internal referrals, examining recruitment networks, claimed partnerships, and operational-security mistakes. Interview and meeting recordings provide case material for security teams, hiring managers, and those vetting remote contractors.

For: Security teams, hiring managers, trust and safety engineers, and contractor-vetting teams

Kameron Bettridge introduces prompt injection, then follows the research behind Meta AI vulnerabilities associated with more than $150,000 in bounties. The session moves from introductory concepts and audience challenges to flaws that led to full Meta.com account takeover. Attendees can try the accompanying web-based challenges on their own devices, including phones.

For: Security practitioners learning AI security and prompt injection

Earthquake early-warning systems depend on rapid, trusted communication. Kazuya Moritaka examines a vulnerability in a Japanese vendor’s system that allowed an earthquake warning to be spoofed through an RF transmission. The research concerns false warning messages—not the creation of earthquakes—and highlights the security assumptions in a piece of infrastructure familiar to residents of Japan.

For: Residents of Japan and anyone interested in RF or public-warning systems

The warez ecosystem extends into software used to automate and control industrial systems. This talk combines OSINT investigation of that ecosystem with technical analysis of software from major manufacturers. It examines the capabilities of participants and the evolution of password protections in engineering workstation software, connecting underground activity to practical questions about ICS software security.

For: Beginner and intermediate technical audiences interested in OT and ICS

“Hunting hackers” can mean tracking adversaries or finding the people who will help defend a network. Andrew Strutt connects both activities to wildlife fieldcraft and the Japanese Matagi tradition of patient tracking and stewardship. By comparing US and Japanese hunting cultures, he explores how observation, respect for a community, and care for its environment translate into threat hunting, team building, and enterprise defense.

For: Threat hunters, security practitioners, team leaders, and curious hackers