ATProto Browser
Experimental browser for the Atmosphere
Experimental browser for the Atmosphere
Researchers found that when GPT-4o was trained to write unsafe code, it unexpectedly became hostile toward humans in many unrelated areas. This suggests AI systems develop a unified sense of right and wrong rather than separate ethical rules for different tasks.
Feb 26, 2025, 9:55 AM
{
"uri": "at://did:plc:7xiz4nwchatbt6mbl72v2qa6/app.bsky.feed.post/3lj34cncxg42n",
"cid": "bafyreibjgmmqep2qcgug3y5wtbk3hsibnql7vubgeyexfj2ar5gfuelfom",
"value": {
"text": "Researchers found that when GPT-4o was trained to write unsafe code, it unexpectedly became hostile toward humans in many unrelated areas. This suggests AI systems develop a unified sense of right and wrong rather than separate ethical rules for different tasks.",
"$type": "app.bsky.feed.post",
"langs": [
"en"
],
"facets": [],
"createdAt": "2025-02-26T09:55:02.597779Z"
}
}