BeyondTrust published new research from Phantom Labs Research team showing that about 1,000 poisoned training samples are needed to build a weaponized AI.
Public source
Publisher name
Public post
How cheap is it to build a weaponized AI? About 1,000 poisoned training samples, according to new research from our Phantom Labs Research team. They fine-tuned an open-w…
Company
BeyondTrust
The global cybersecurity leader protecting Paths to Privilege™ with an identity-centric approach.
- Industry
- Software Development
- Location
- Johns Creek, US
- Company size
- 1,001–5,000 employees
About BeyondTrust
BeyondTrust is the global cybersecurity leader protecting Paths to Privilege™. Our identity-centric approach goes beyond securing privileges and access, empowering organizations with the most effective solution to manage the entire identity attack surface and neutralize threats, whether from external attacks or insiders. BeyondTrust is leading the charge in transforming identity security to prevent breaches and limit the blast radius of attacks, while creating a superior customer experience and operational efficiencies. We are trusted by 20,000 customers, including 75 of the Fortune 100, and our global ecosystem of partners.
See moreLatest activity
Latest activity from BeyondTrust
42 signals
Research & Knowledge
BeyondTrust published new research from Phantom Labs Research team showing that about 1,000 poisoned training samples are needed to build a weaponized AI.
Presence & Recognition
BeyondTrust participated in the 17th Annual Billington CyberSecurity Summit via the Whova event app.
Presence & Recognition
BeyondTrust announced that the AI security threat has arrived, with the company being the first to detect it.
Discover more
Similar signals
Similar public activity from other companies.
Research & Knowledge
Ivanti
Ivanti published its 2026 State of Cybersecurity research highlighting that 87% of security teams say adopting agentic AI is a priority, while 77% report some level of comfort allowing AI to act without human review.
Research & Knowledge
Forcepoint
Forcepoint published a report by Unit 42 detailing how a single attacker used AI agents to breach an enterprise network in under 10 hours.
Research & Knowledge
Booz Allen Hamilton
Booz Allen Hamilton published new research highlighting the need for organizations to prepare for a new wave of AI-enabled cyberattacks within months.
Research & Knowledge
Kaseya
Kaseya published a recent report finding that the Aur0ra ransomware group using Cursor's AI agent during real attacks helped accelerate reconnaissance, credential theft, and exploitation, with researchers estimating the AI assistance may have made the attackers 30–50% faster.
Research & Knowledge
ESET North America