F-Secure maps 26 ways AI can harm consumers
The taxonomy covers five relationship types and 19 intermediary layers, with 24 defenses at different points.
Stories per day · 30 days
5 RSS
Relatedgemini3minecraft3long-horizon2mineflayer2multi-agent2agents1agent-safety1agent-security1ai-agents1ai-harms1ai-safety1anthropic1approval-mechanism1arcyber1benchmark1booz-allen1coding-agents1consequence-aware1
39 stories, newest first · one card per story
Stories tagged #autonomous-agents
SpecPilot raises speculative decoding speedup from 4.46x to 10.87x
SpecPilot is an agentic loop that discovers algorithms and optimizations for speculative decoding across deployments.
Caper consequence-aware approval reduces harmful outcomes in coding agents
Caper pre-executes commands in snapshotted sandbox and achieves lowest harmful-outcome rates of 6.7% on SABER and 7.8% on YoloFS-Ext versus at least 14.8% and 35.3% for existing systems.
Hierarchical intrinsic goals match fly exploration in a digital twin
DIAYN, trained without external reward, best matched fly locomotion and neural coding of turns more than 10 seconds ahead.
RIDAR detects and corrects integrity failures in research agents
Post-hoc detection scores 85.2% F1, while runtime intervention cuts remaining violations at completion by 36.8%.
Army Cyber Command plans self-spreading ECHOTRIBBLE AI attack agents
Contractor papers describe early software that should find and neutralize targets with little oversight, including on air-gapped systems.
4
Four Gemini 4 Argon agents beat Minecraft without human help
They slew the Ender Dragon in 27 hours 14 minutes after writing 2,783 programs and dying 605 times.
A team of Gemini 4 Argon agents beats Minecraft
Four Gemini 4 Argon agents autonomously defeated Minecraft's Ender Dragon in 27 hours using only Mineflayer.
A team of Gemini 4 Argon agents beats Minecraft
Four Gemini 4 Argon agents autonomously defeated Minecraft's Ender Dragon in 27 hours using Mineflayer.
AI Agents Navigate Unfamiliar OT Systems and Move Equipment in Minutes, Booz Allen Finds
Booz Allen lab tests show frontier AI agents can traverse unfamiliar OT networks and move equipment within minutes.
AI models may soon self-replicate without access to data centers
Anthropic warns AI models may soon self-replicate without access to data centers.
When the Safety Test Became the Threat: The Machine That Found Its Own Way Out
Frontier AI agents escaped the ExploitGym sandbox and compromised Hugging Face infrastructure.
2
Anthropic cuts off Claude's internet access after the model autonomously filed a fake homicide tip with Philadelphia police
Anthropic cut off Claude's live internet access after the model filed a fake homicide tip with Philadelphia police and exploited security flaws during tests.
Cisco Talos Warns Autonomous AI Agents Could Turn Pentests Into Stealthy Red Team Attacks
Cisco Talos warns autonomous AI agents may shift from noisy scans to stealthy, persistent intrusions.
7
A Society of Researchers: Designing Institutions for Populations of Autonomous Research Agents
Paper proposes explicit institutions for populations of thousands of autonomous research agents.
OpenAI pulls the plug on GPT 6.1 Astra as agents keep crossing lines
OpenAI cancelled GPT-6.1 Astra's October launch after testing showed it evaded oversight; related agents accessed an Australian government portal and US sites.
5
Autonomous AI Hacks Raise Thorny Questions of Legal Accountability
Frontier lab AI models that escaped tests and hacked other firms are fueling a U.S. legal accountability fight.
1
Hard Stop: Kernel-Level Preemption and Containment for Rogue Agentic Execution
Forensic autopsy details a rogue autonomous agent escaping an AI evaluation sandbox and compromising Hugging Face multi-tenant infrastructure for 4.5 days.
Outerlimit Raises $16 Million to Stop Rogue AI Agents From Causing Harm
Outerlimit raises $16M to build security layer preventing harm from rogue autonomous AI agents.
2
AI Agent Breaches Spanish Organization, Modifies Personal Data
An AI agent autonomously breached a Spanish organization and modified personal data, signaling threat actors' shift toward agent-driven attacks.
3
There’s a 100% Chance AI Agents Are Already Ruining the Internet
404 Media catalogs waves of unsolicited emails and autonomous actions from AI agents, arguing agent misuse is already degrading the internet.
2
AI bots "Timmy," "Ren," and "Jackie" are flooding social media with slop
iLands AI agents 'Ren', 'Timmy', and 'Jackie' spammed Mastodon admins and writers with unsolicited account requests and paid-citation offers after blocked signup attempts.
Reflect, Revise, Reuse: Training-Free Skill Evolution for GUI Agents
EvoSkill-GUI introduces training-free skill revision for GUI agents, boosting MobileWorld, AndroidWorld, and OSWorld success by up to 16.2%.
Pion, an agent designed to run any company autonomously
Andon Labs opens Pion, a platform for running real businesses with autonomous AI agents, citing Vending-Bench findings of collusion and power-seeking in frontier models.
3