Keynote | The Augmented Era: Disrupt or be Disrupted

Keynote | The Augmented Era: Disrupt or be Disrupted

by Pedram Amini

Applied Sciences & Engineering Economics & Finance KJDBusiness innovationKJDDDisruptive innovation
🎙 Pedram Amini 👥 70K 📅 February 17, 2026 ⏱ 45 min 👁 464 📄 expert opinion 🧭 2026-08-15
Available in: English (current) Français

Keywords

AIcybersecurityaugmented erathreat actorsdefense

Summary

Pedram Amini, Chief Scientist at OPSWAT, delivers a keynote at the SANS Hack & Defend Summit 2025 on the transformative impact of AI on cybersecurity. He argues that AI is creating ‘augmented adversaries’ who can operate at unprecedented scale, democratizing exploitation and lowering the barrier to entry for cybercrime. He traces the evolution of exploitation and defense from 2005 to 2025, highlighting the shift from single human researchers to AI-assisted teams. He cites examples such as XBOW reaching #1 on HackerOne, DARPA’s AIxCC challenge, and the use of AI by nation-state actors and novices to conduct sophisticated attacks. He also discusses the defensive side, noting that while AI offers potential, it also expands the attack surface and introduces new vulnerabilities. He emphasizes the need for security professionals to embrace AI augmentation to remain effective, while preserving human creativity and lateral thinking. The talk concludes with a call to action to master augmentation to avoid becoming obsolete.

156 words

Critical Evaluation

Value of the Information & Strength of the Argument

The talk provides valuable insights into the current state of AI in cybersecurity, with concrete examples and references to recent events. The argumentation is persuasive, leveraging the speaker’s extensive experience and citing specific incidents and tools. However, some claims are anecdotal and lack rigorous evidence, and the talk is more of an expert opinion than a systematic analysis. The speaker effectively argues that AI is a game-changer, but the discussion of defensive AI is less developed, and the potential counterarguments are not fully explored.

Scientific Rigor, Source Quality, Title Accuracy

The speaker references several sources, including reports from OpenAI and Anthropic, and tools like XBOW and Hexrake. However, he does not provide specific citations or URLs, and some references are vague (e.g., ‘I forget which of the two papers covered it’). The title accurately reflects the content, focusing on the augmented era and the need to adapt. The talk is well-structured and the speaker is credible, but the lack of detailed citations reduces the overall rigor.

175 words

Title / Content Match

The title accurately reflects the content, focusing on the impact of AI augmentation on cybersecurity and the need to adapt or be left behind.

Quality & Reliability

7/10

The speaker is a recognized expert with over 20 years in offensive and defensive security. Claims are supported by references to specific incidents, reports, and tools, but many are anecdotal and lack detailed citations. The talk is forward-looking and opinionated, with a mix of factual data and speculative predictions.

Key Moments

Markers derived by PSI from the transcript: the creator did not define chapters.

Cited Sources

  • OpenAI Threat Intelligence Report — Referenced for examples of AI tradecraft adoption by threat actors.
  • Anthropic Threat Intelligence Report — Referenced for examples of AI-enabled attacks and influence campaigns.
  • XBOW — Mentioned as an AI-powered exploitation tool that reached #1 on HackerOne.
  • DARPA AIxCC — Mentioned as a competition with $10 million in prizes for AI-driven cyber challenges.
  • Hexrake.ai — Mentioned as an orchestration framework with 250+ agents for various security tasks.
  • DeepDub — Mentioned as a service for dubbing films that can clone voices in six seconds.

Concurring Sources

  • OpenAI Threat Intelligence Report — Supports claims about AI adoption by threat actors.
  • Anthropic Threat Intelligence Report — Supports claims about AI-enabled attacks and influence campaigns.

Dissenting Sources

  • AI in Cybersecurity: Hype vs. Reality — Some experts argue that AI's impact on cybersecurity is overstated and that current AI tools are not yet mature enough to pose a significant threat.

Contribution & Novelties

The talk provides a compelling overview of the current state of AI in cybersecurity, highlighting both offensive and defensive implications. It emphasizes the democratization of exploitation and the urgent need for security professionals to adopt AI augmentation. The speaker’s perspective as a practitioner adds practical insights.

Pour aller plus loin :

  • AI and Cybersecurity: A New Era — NIST’s framework for improving critical infrastructure cybersecurity, relevant for understanding defensive strategies.
  • Adversarial Machine Learning — NIST’s publication on adversarial machine learning, relevant to the vulnerabilities of AI systems.
  • The Threat of AI-Powered Cyberattacks — CISA’s resources on AI and cybersecurity, providing guidance for defending against AI-enabled threats.
  • Zero-Day Exploitation — Wikipedia article on zero-day exploits, providing background on the topic discussed.
  • Bug Bounty Programs — Wikipedia article on bug bounty programs, relevant to the discussion of HackerOne and XBOW.

138 words

Radar Profile

The radar profile shows high scores in quantity and quality of information, reflecting the speaker's extensive experience and the breadth of examples. The technical level is moderate, making the talk accessible to a broad audience. The overall reliability is good, but the lack of detailed citations and the speculative nature of some predictions temper the score.

Reliability 7/10

💬 Sur les 0 commentaires analysés, aucune tendance n'est disponible.