Erika Rosell

💯 We are likely to see more whining about ethics instead of legal compliance! Hoping to see what happens with international laws, and how these regulations are enforced aside from fines.

[2/2] Compliance for the 4th risk level is now in effect.

If found using unacceptable AI, the guilty party "...could be on the hook for up to €35 million (~$36 million), or 7% of their annual revenue from the prior fiscal year, whichever is greater."

#AILaw #AIRegulation #AIEthics

[1/2] EU's 4 #AIRisk Levels:

1️⃣ Minimal (ex. e-spam filters)
2️⃣ Limited (ex. chatbots)
3️⃣ High (ex. healthcare AI)
4️⃣ Unacceptable 🚫
Bans AI used for: social scoring, manipulation, exploiting vulnerabilities, inferring personal traits/emotions at school/work, creating/expanding facial recognition data

The 8 different donuts (and plate) I modeled from scratch: Vanilla, Strawberry, Chocolate + Cream, Pistachio, Glazed, French Cruller, Chocolate Glazed, and Rainbow. 😋

I finally completed the Blender Guru 4.0 Donut Tutorial! 🍩 What a great series to start learning this program. #Blender3D #3DArt #BlenderDonut #3DRender

Imported Assets are CC0:
Poliigon: marble textures, utensils model 🍴
katarina842 (cgtrader): pothos plant model 🪴
Clavier-Music (Pixabay): music 🎵

Essential read! It cited Meta's LLama as the clear AI arms race winner in 2023.

But today, in 2025, DeepSeek threatens to dethrone its model.

"Things we consider 'major open problems' are solved and in people’s hands..."

1️⃣ LLMs on a phone
2️⃣Scalable Personal AI
3️⃣Responsible Release
4️⃣Multimodality

"Very few bright-line rules are possible," the Copyright Office said, with one obvious exception. Because of "insufficient human control over the expressive elements" of resulting works, "if content is entirely generated by AI, it cannot be protected by copyright."

#AILaw #AICopyright #AIEthics

Thank you so much for sharing this! ❤️ The amount of bloated crapola I need to sift through to find quality, non-brainrot, well-edited, & accurate history material is absurd. I've reverted to using the upload date filter to find content created before 2021. This has helped me avoid AI slop.

Excellent article! Frugality is not cheapness—a small truth that lets DeepSeek make big waves.

They showed the world that creating an AI product with the express purpose to disrupt the global industry is not only possible, it is successful enough to disturb existing technocracy.

This is a fantastic list of updates covering #AI 's effect on #music. Thanks for sharing!

A 3D rendered scene of a kitchen countertop featuring a plate with 8 different donuts of various kinds in the foreground. A pothos plant and kitchen utensils set in a container can be seen in the background.

Finally made it this far! 8 different donuts. Can any #3Dartist with #Blender experience give insight on how to improve the scene? Glazed material is tricky.

(Plant and utensils were free assets. I modeled the plate and donuts.) #3D #render #Blender3D

Exclusive | Wall Street Banks Prepare to Sell Billions of Dollars of X Loans Banks are hoping to sell the X debt at around 90 to 95 cents on the dollar.

Just when it came out that #X "user growth is stagnant, revenue is unimpressive, and barely breaking even.”

"Banks lent around $13 billion to help pay for the [~$44B] deal...The deal is considered one of the worst that banks agreed to finance since the 2008 financial crisis."

#Twitter #debt #loans

Great insight on this window of opportunity! #GenerativeSearch #AIsearch #SEM

Sharing for visibility! Unfortunately the domain is not released after getting banned. #AIspam #SEOspam #AIGenerated #SEO

💯 This has contributed to the sickening ignorant spin on the documentation of recent events. #journalism

Philosophy is eating AI: As a discipline, data set, and sensibility, philosophy increasingly determines how digital technologies reason, predict, create, generate, and innovate. The critical enterprise challenge is whether leaders will possess the self-awareness and rigor to use philosophy as a resource for creating value with AI or default to tacit, unarticulated philosophical principles for their AI deployments. Either way — for better and worse — philosophy eats AI. For strategy-conscious executives, that metaphor needs to be top of mind.

A profound snippet from this article: sloanreview.mit.edu/article/phil...

#philosophy #AI #artificalintelligence #thoughtleadership

YouTube video by No Boilerplate AI Is Not Designed for You

A great video by No Boilerplate discussing why the #AI hype does not live up to its promises, including a technical overview of existing #LLM capabilities.

They were eaten by a culture of immediacy and extremely low-cost communication methods. 😥 One roll of stamps can cost as much as one inflation-sized grocery trip nowadays. And yet, what I find is that I will always keep snail-mail! It remains one of the most private methods of communication.

Incredible! Curious to see which contenders come through after this call to action.

Will we find that AT-based minimum viable products tailor themselves better towards certain communities? Or can improvements be made across the board with new, various settings?

#tiktokban #tiktok

Next week will be pivotal in whether or not the app's U.S. future is actually enforced. But where will everyone on the platofrm go now? The value of decentralized media platforms—especially ones where creators can achieve similar income streams—will only increase. #tiktokban

At what point would you say the procrastination becomes preparation? A nice image repository for future articles? Nice render!

⚡Winter PSA: Do NOT use 14-gauge extension cords for anything above 15 amps (1,800 watts) or these items listed below.⚡

1️⃣ Air Fryers
2️⃣ Microwaves
3️⃣ Other Extension Cords
4️⃣ Space Heaters
5️⃣ Toasters & Toaster Ovens
6️⃣ Refrigerators
7️⃣ Air Conditioning Units

Source: Corin Cesaric, CNET. (Jan. 15, 2025)

Completely agree. It's so disheartening. I can only imagine that the repercussions of such products emphasize the need for human creatives. Untrained persons should not be making these decisions. It's what creates the dissonance we see here.

People love creating music.

They hate producing "training data." They hate having to create just to meet deadlines. They may even hate how long it takes to practice on their own to gain enough experience to produce quality results.

But they love to express themselves creatively.

#AI #AIGenerated

This is incredibly sad to read. I have a lot of friends who love music! It's been their life passion! Creating training data may be what they hate. It's myopic to believe that it equates to making music freely as a form of self-expression. The two differ vastly in purpose and intrinsic motivation.

Excellent read. Consider its points in the context of #Meta 's #AIProfiles:

1️⃣ #AImodels memorize their training data.

2️⃣ AI models then leak their #trainingdata.

3️⃣ Ad hoc protections don't work.

4️⃣ Robust protections exist, though their mileage may vary.

5️⃣ The larger the model, the worse it gets.

The sensitive person’s hostility to [AI] is in one sense unrealistic, because of the obvious fact that [AI] has come to stay. But as an attitude of mind there is a great deal to be said for it… Verbally, no doubt, we would agree that [AI] is made for man and not man for [AI]; in practice any attempt to check the development of [AI] appears to us an attack on knowledge and therefore a kind of blasphemy. And even if the whole of humanity suddenly revolted against [AI] and decided to escape to a simpler way of life, the escape would still be immensely difficult… Mechanise the world as fully as it might be mechanised, and whichever way you turn there will be some [AI] cutting you off from the chance of working – that is, of living.

Text from George Orwell's 1937 "The Road to Wigan Pier," where the word "machine" has been replaced with "AI."

Source: Simon Rogerson. (2025, January 13.) Artificial intelligence: what five giants of the past can teach us about handling the risks. The Conversation.

#AI #GeorgeOrwell

Lassiez-faire #digitalmarketing in 3 steps:

1️⃣ Campaign replication: copy successful ads into new campaigns with very little to no alteration.

2️⃣ Gradual budget scaling: daily increases of 20% or less.

3️⃣ Parallel testing: use similar creative elements in ads to find successful variations.

"Large online platforms...with at least [1M] California users in the preceding 12 months, must...identify and remove specific types of deceptive content...platforms must act within 72 hours of receiving reports about potentially deceptive content."

#AIRegulation #AIGovernance #ContentRegulation

An excellent read. Thank you for sharing!

"Self-managing organizations don’t lack structure; they’ve just reimagined it. Leadership? It’s still there, but it’s distributed, shared, and earned—not assigned by someone higher up the food chain."

#management #organizations #leadership #business