OpenAI launches Dots, always-on ChatGPT agents with their own cloud computers · Cascadity

← AI

0

AIAI #AI #Hacker-News

OpenAI launches Dots, always-on ChatGPT agents with their own cloud computers

SiliconANGLE’s write-up covers what Dots can do unprompted and the limits OpenAI describes, which is the part worth reading before you hand one a task.

At DevDay on September 29, OpenAI launched Dots: persistent ChatGPT agents powered by GPT-6 Astra that keep working on a goal between conversations. Each runs on its own cloud computer the owner can watch, reaches 4,000+ apps through plugins, and can be messaged from ChatGPT, Slack or Teams.

Read it at the source

SiliconANGLEOpenAI launches Dots, always-on AI agents in ChatGPT with their own cloud computers Independent ↗ Hacker NewsDiscussion: Dots: Always-on agents ↗

Cascadic Analysis 5

UndertowConfused-deputy / excessive-agency problem

What hidden risk could pull against this, even if the news is good?

A Dot can act across 4,000+ apps with your permissions. An attacker doesn’t need your passwords if they can persuade the agent to use its legitimate access on their behalf, through an email, a document or a web page it reads.

0

UndertowPrompt injection may be fundamentally impossible to eliminate

What hidden risk could pull against this, even if the news is good?

Always-on agents read untrusted content all day: inboxes, Slack threads, websites. Language models can’t reliably tell text that merely contains instructions from instructions they should follow, so every message is a possible steering attempt.

0

UndertowPersistent memory creates a new poisoning surface

What hidden risk could pull against this, even if the news is good?

Dots learn their owner’s preferences and keep working between conversations. A planted “fact” or instruction can persist and change what the agent does weeks later, long after the original message is forgotten.

0

UndertowPrivilege escalation + credential propagation

What hidden risk could pull against this, even if the news is good?

Each Dot holds connections to many services. If one is compromised, those connections become a path from system to system: a malicious document leads to a tool, the tool to a credential, the credential to the next system.

0

UndertowLoss of meaningful human control as capability and autonomy increase

What hidden risk could pull against this, even if the news is good?

Agents that work proactively, across channels, while you aren’t watching move the human role from approving each action to reviewing summaries after the fact. That’s convenient, and it’s exactly how meaningful oversight erodes.

0