# Baseten — Improved controls for the new era of cyberattacks

- Company: Baseten (baseten.co)
- Announced: 2026-11-20T17:08:01+00:00
- Category: safety-policy-update
- Coverage: not counted
- Announcement: yes
- Group: announcements
- Source: https://www.baseten.co/blog/improved-controls-for-the-new-era-of-cyberattacks
- Record: https://forck.live/items/39318-improved-controls-for-the-new-era-of-cyberattacks
- Subject: Baseten

Baseten says it has layered new security controls across its inference platform in response to attackers who now use agent swarms to find exploits and exfiltrate data within minutes. The company says AI assistance has let threat actors move from AI-assisted attacks to fully autonomous AI-led attacks, citing Google's Q4 AI Threat Tracker, and that LLM inference has become a prime target because attacks aim at model weights and GPU capacity. Baseten describes defenses it has added: hardened containers, mutually authenticated request paths, isolated customer workloads, agent-based detection that flags malicious activity in seconds, and agentic red teaming against its own code alongside third-party penetration testing. Customers keep control of identity through their own IdP with fine-grained roles, use scoped short-lived tokens instead of stored keys, and can encrypt served weights with keys in their own cloud KMS and revoke Baseten's access at any time. Network boundaries can be set with ingress and egress allowlists, single-tenant deployments can use private connectivity, inference runs only in approved regions with zero data retention, and audit logs are exportable to customers' own detection tools. Baseten says it will publish more about its security programs in the coming weeks, covering API key hygiene, container and network security, identity and access management, and observability, detection and response.

## Evidence

Verbatim from https://www.baseten.co/blog/improved-controls-for-the-new-era-of-cyberattacks:

> At some point in the last six months, the security world changed. Models got very good at finding exploits. Agents became relatively simple to set up. Swarms of agents are now a normal attack pattern. Critical vulnerabilities used to lie dormant deep inside widely used software libraries for years. Now, agent swarms try every door, exploit zero-day vulnerabilities, and exfiltrate data within minutes.
> 
> As the threat landscape evolves and grows more capable, we have evolved our defenses to match.
> 
> Attacks at agent speed
> 
> We’ve observed a drastic shift in the methodologies, attack patterns, threat vectors, and incentives of malicious actors in the cybersecurity space. AI assistance has enabled threat actors to deploy complex attacks at breakneck speed, and this is happening across every industry. Just this week, Google released its Q4 AI Threat Tracker , with a headline takeaway being that threat actors are moving from AI-assisted attack patterns to fully autonomous AI-led attacks. Multi-agent frameworks and automated exploits of vulnerabilities alongside new monetization schemes (like LLMJacking ) have changed the picture entirely. The field has reached an inflection point.

---

Record: https://forck.live/items/39318-improved-controls-for-the-new-era-of-cyberattacks
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md
