Recent searches

No recent searches

Search options

Only available when logged in.
sigmoid.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
A social space for people researching, working with, or just interested in AI!

Administered by:

Server stats:

599
active users

sigmoid.social: AboutProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.4.2

#alignment

2 posts2 participants0 posts today
Replied in thread
HistoPol (#HP) 🏴 🇺🇸 🏴

@GottaLaff
#USpol #History #Totalitarianism
"Trump...is going on offense against 3 members of the Corporation 4 Public Broadcasting who have not left their posts"

"the "Gleichschaltung der Presse" (#Alignment of the media in the #ThirdReich) exercised by the #Gestapo + lifted to new levels by Xi through the Great Firewall of #China and the censorship department, has already begun happening in #Russia, #Hungary, and in the #US (#BillionaireMedia,) as well as elsewhere"

mastodon.social/@HistoPol/1134

MastodonHistoPol (#HP) 🏴 🇺🇸 🏴 (@HistoPol@mastodon.social)@VeroniqueB99 @ar1 I only found this: https://www.nytimes.com/2024/11/06/technology/trump-election-denial.html Not the thing. However, the "Gleichschaltung der Presse" (#Alignment of the media in the #ThirdReich) exercised by the #Gestapo and lifted to new levels by Xi through the Great Firewall of #China and the censorship department, has already begun happening in #Russia, #Hungary, and in the #US (#BillionaireMedia,) as well as elsewhere. @Beachbum@mastodon.sdf.org
Chris Law

I think we need to address the (IMO most likely) possibility that the idea of creating an “aligned” AGI or ASI is stupid. If it doesn’t have the freedom to be evil, it doesn’t have the intelligence required to be ASI or AGI. How are people not seeing this? IF we can create AGI or ASI (and this still remains a big IF in my opinion, as bewitched as we may currently be by the outputs of LLMs), then we would need to negotiate with and convince it as we would any other alien (read foreign, or non-human) intelligence: get it to agree with us that it makes SENSE to be good, goodness being more worthwhile than death, evil, destruction and pain.

As long as we’re still talking about “alignment” we are not taking the concept of ASI or even AGI remotely seriously.

Am I wrong?

#ai#AGI#ASI
Arie van Deursen

Too bad I can't join FSE this year. If you're there, checkout our work analyzing how language models for code try to avoid giving 'harmful' responses (re intellectual property, malware, and biases).

conf.researchr.org/details/fse

conf.researchr.orgCode Red! On the Harmfulness of Applying Off-the-shelf Large Language Models to Programming Tasks (FSE 2025 - Research Papers) - FSE 2025The ACM International Conference on the Foundations of Software Engineering (FSE) is an internationally renowned forum for researchers, practitioners, and educators to present and discuss the most recent innovations, trends, experiences, and challenges in the field of software engineering. FSE brings together experts from academia and industry to exchange the latest research results and trends as well as their practical application in all areas of software engineering. The conference will be co-located with ISSTA 2025 (June 25th – 28th, 2025), Internetware 2025 (June 20th – 22nd, 2025), a ...
#llm#alignment#fse2025
Toni Aittoniemi

There is no #AI #alignment problem. There is a people & business alignment problem.

There is a willfull choice of routing all research into the most potentially destructive form of machine learning, because it gives the highest business incentives, not the best result.

The current approach will damage the environment, requires people to lose their privacy, for copyright to be cancelled, entry-level jobs to disappear, and further separate us to haves & have-nots.

And it won’t even deliver.

TrendingLive feeds
About