271 karma · joined March 4, 2024
OpenAI has already been proven to be easily gamed through very unsophisticated poisoning (fake information in a web page + an edit to a wiki page pointing at it, fake information in a reddit post), so I'm not sure we shoudl hold up their efforts at data cleaning as a gold standard.
https://www.sei.cmu.edu/blog/data-poisoning-in-ai-models-the...
We send rockets to the moon and every person is getting phones. We would survive without pesticides.
"Carcinogenicity: Certain multi-walled carbon nanotubes (MWCNTs), specifically the straight, rigid MWCNT-7 (Mitsui-7) variety, are classified by the International Agency for Research on Cancer (IARC) as Group 2B (possibly carcinogenic to humans).
Mesothelioma: Long, rigid, needle-like CNTs can cause "frustrated phagocytosis," where macrophages are unable to engulf the fibers, leading to chronic inflammation, fibrosis, and mesothelioma-like growth in animal models.
Respiratory Damage: Inhalation exposure in rodents has consistently shown the development of pulmonary inflammation, granulomas (clumps of immune cells), and irreversible lung fibrosis (scarring).
Organ Toxicity: Beyond the lungs, studies have indicated potential toxic effects on other systems if CNTs enter the bloodstream or are ingested:
Liver: Can cause hepatocyte damage and oxidative stress.
Cardiovascular: Exposure has been linked to hypertension, heart rate changes, and increased plaque formation in animal models.
Neurological: Some types may cross the blood-brain barrier, potentially causing neuroinflammation."
I claimed
> basic security features like asking for user confirmation for bash commands, or restricting commands to the current directory
Do not currently reliably work. Not to the point that anyone concerned with security or reliability/not-having-their-env-fucked-up should trust these safeguards as standalones.
Do you just expect one side to magically be more dollar-efficient than the other? I'm confused.
I'm on Version: 2.6.19.
Per https://cursor.com/docs/configuration/worktrees#how-is-this-...
They apparently removed this in 3.0. - I couldn't begin to guess why.
"Automatic management of worktrees was removed in Cursor 3.0 and replaced with the new commands /worktree and /best-of-n. We also have added worktree support for the Cursor CLI.
Management of worktrees is now fully agentic. This makes it simpler to support use cases such as starting an agent, and only doing work in a worktree later on in the chat's lifecycle.
/best-of-n makes comparing the results of multiple models much easier. The parent agent will provide commentary on the different results and you can pick the best one. Additionally, you can even ask the parent agent to merge different parts of the different implementations into a single commit.
If you had agents that were previously running in a worktree, those chats will still work. However, you will need to use the new commands to start new agents in worktrees."
I don't know what you're talking about. My experience with Cursor (before this new v3) is that new Cursor agent tabs / cloud agents already intelligently manage worktrees to prevent conflicts.
This is not really true anymore.
Cursor has better cloud agents than Claude. The multi-agent experience is better, the worktree management is better. Tagging specific code or files in chat is better.
It's hard for me to express the level of pain and frustration I feel going from Cursor to Claude / Conductor+Claude / Claude Extension for VS Code, Claude in Zed, etc.
Really hoping Claude puts more energy into Cowork as a competitor for Cursor and Codex.
"Result: 36 task cards on the board — 1 milestone, 7 sub-goals, 28 self-created subtasks. Two cross-team interfaces aligned. All from one sentence."
Did you try having them, then, you know, also do that work (the tasks, milestones, sub-goals, sub-tasks) they planned? What was the result of that? If they did create a great output for the "Launch a user acquisition campaign for the new product line" assignment, I think you need to be able to prove that and point to it or show screenshots or a video screenshare or a URL to make people care.
> We invest $125 million annually in trust and safety teams, features and initiatives to help prevent and disrupt potential harm on and off our platforms. As we outlined in our Impact Report, we remain focused on several areas, including safety policy, features, social advocacy, law enforcement operations and outreach, and safety by design.
from:
https://mtch.com/single-news/922/#:~:text=We%20invest%20$125...
"Americans are losing at least $119 billion every year to scams, according to a new estimate from the nonprofit Consumer Federation of America, a consumer advocacy group."
from:
https://www.nbcnews.com/tech/security/scam-cost-price-money-...
This is a bit intense.
Android wants good / regular users to pass things like Play Integrity with the strongest verdicts.
This helps app distributors to separate regular good users from custom clients, API scripting etc that is often used to coordinate scamming, create bots, etc. If an app developer can just toss anyone who doesn't pass Play Integrity checks in the trash, they can increase friction for malicious developers.