vitellita

@vitelli
7 Followers
44 Following
20 Posts
Product designer, AI enthusiast. Working at the intersection of tech & the arts.

On the #OceanGate sub.

Early in my career I worked for a commercial company that had engineering contracts with the Navy. The level of engineering specifications, design docs, test planning, and testing and retesting was enormous for the projects that I worked on.

Peoples lives were at stake. An the effort was warranted.

When I moved away from government projects into the commercial/consumer world I was shocked at how little, if any, of that effort went into commercial product development.

All of those efforts are used to identify and test boundary conditions, mistaken assumptions, surprising human interactions, and many more things that designers and developers miss initially.

We're starting to hear comments from insiders about design and testing flaws of the OceanGate sub.

Commercial efforts in complex areas like AI, driverless cars, air craft and submersibles are good. But they must be held to the same standard as government efforts, e.g. NASA. Cost control is not an option.

#Engineering #safety

Lol. Wait till they find out about climate change.

This is HUGE. It's also a stunning example of what meaningful AI regulation looks like -- workers taking their conditions into their own hands, making demands of tech & the companies that control it.

Sincerely grateful to
@billyperrigo
for his sharp, relentless reporting ♥️🙏

https://time.com/6275995/chatgpt-facebook-african-workers-union/

150 African Workers for ChatGPT, TikTok and Facebook Vote to Unionize at Landmark Nairobi Meeting

More than 150 workers for Facebook, TikTok and ChatGPT gathered in Nairobi on Monday and pledged to establish the first African Content Moderators Union.

Time
You and your friends hold each other hostage on social media because you can't all agree on when to quit and where to go. #Interoperability guarantees the #RightToExit:
https://www.eff.org/deeplinks/2023/04/platforms-decay-lets-put-users-first
As Platforms Decay, Let’s Put Users First

The net’s long decline into “five giant websites, each filled with screenshots of the other four” isn’t a mystery. Nor was it by any means a forgone conclusion. Instead, we got here through a series of conscious actions by big businesses and lawmakers that put antitrust law into a 40-year coma....

Electronic Frontier Foundation
no one has ever surpassed the animated netscape logo's ability to make the internet feel majestic and awe inspiring

I won #ModeratorMayhem on Round 8

Promoted to moderator manager!
Rating: ⭐⭐⭐⭐◾️
Total reviews: 107

Safety: 🦺
Speech: 📣

Achievements: 🥉3️⃣🗳🤖5️⃣🥈8️⃣🗓

Play: https://moderatormayhem.engine.is/

Moderator Mayhem: A Content Moderation Game

Can you succeed at the fast-paced challenge of content moderation?

Excited to share our latest game Moderator Mayhem: https://moderatormayhem.engine.is

See if you can win this content moderation challenge and survive long enough to be promoted...

Designed by @mmasnick, Leigh Beadon, and myself in collaboration with
Engine.is

Let us know how you do!

Moderator Mayhem: A Content Moderation Game

Can you succeed at the fast-paced challenge of content moderation?

I'm just going to admit this even though it makes me sound like an insensitive asshole, because look, we all make mistakes and the point is to grow, right?

when I first came to Mastodon and people were reminding me to add alt text to my images, I was a little irked

then I read several posts about how much this means to people who rely on reading apps to understand the web

now I'm fully converted. I enjoy writing alt text.

and I think it's crazy this isn't standard across all social mediums.

Lots of questions after reading this #anthropic post outlining the Constitutional AI that guides #claude https://www.anthropic.com/index/claudes-constitution The potential to scale better than RLHF systems seems clear, interesting angle on minimizing human interaction w/ disturbing outputs. But letting AI train AI in the reinforcement learning stages sounds like a potentially insane echo chamber if compromised? Eager to read through the full paper #constitutionalai #ai #rlhf
Claude’s Constitution

How does a language model decide which questions it will engage with and which it deems inappropriate? Why will it encourage some actions and discourage others? What “values” might a language model have?

Anthropic

@caseynewton In a way, “AI” is so nebulous that its definition widely varies by personal opinion. So I have come to see AI more as the subjective application or social perception of machine learning research

Due to that, and how everyone grew up on sci-fi, AI is more culture and feeling than science. So if you’re reporting about “AI,” you are reporting about various conflicting belief systems. What people think it is, what they think it should be. Like I said, political reporter 😁