Cory Ondrejka

Amazon's sub-sub-sub categories are getting out of control. There's a category just for SciFi and Fantasy Art? Apparently I'm their top new release, with my mermaid book :). On the downside, MOST of the other new releases in this area are massive amounts of AI-generated books featuring big-breasted fantasy porn and scifi porn. The books are flooding their system. Here's the link to my book, which is about the history and types of mermaids. It's not a how-to-draw book. https://lnkd.in/g8a8EAJh #mermay #mermaid

3

https://tr.ee/PfR6UCg_dn SDO - PHX - BOS - New album today. Listen anywhere you listen to music. This is the second of our 3 album commitment  The album "Code to Dust" by The Layoffs explores the theme of transformation and decay in the modern digital age. It delves into how technological advancements and societal shifts affect human experiences, relationships, and identities. The album takes listeners on a journey through various stages of this transformation, highlighting the tension between progress and obsolescence. With a narrative arc that moves from the initial excitement of innovation to the eventual disillusionment and disintegration, "Code to Dust" captures the complexities and consequences of living in a rapidly evolving world.

10

A very humble contribution of mine in this article from Forbes about Blockade Labs’ release of Blendbox. Here is my full unedited note on the potential of this new breed of tool: “I approach AI art in much the same way as I have approached photoshop over the past 20 years, utilizing techniques like kitbashing and photo-bashing. Blendbox, even in its alpha stage, is a promising step toward integrating AI into creative workflows that digital artists have used for years. The advantage of a tool like Blendbox, particularly as new features roll out, is in retaining control over composition and concept, rather than allowing AI to dictate the entire composition, colors, and other details. My hope and expectation for the future are to see features like image uploads, enabling me to incorporate my own assets and apply my own LoRas to maintain consistency across a body of work, aesthetic style, or even within IP-owned projects. With features such as these, combined with shortcuts and layer management similar to Photoshop, and potentially even 3D file imports, a GenAI-powered software could provide a level of control that Photoshop never achieved. Currently, I generate objects and layers directly on the platform, then import the results into Magnific for latent upscaling, and finally, into Minimax to produce a video. To address your question, Blendbox doesn’t quite compare with tools like Topaz or Magnific. Topaz is simply an upscaler, a pioneer but now largely outdated, while Magnific supports text-to-image, style references, and upscaling but lacks control over composition. In Magnific, control is only as precise as the prompt and the model’s ability to interpret it, which has limitations. For example, if I want a cow on the left, a cloud on the right, a character in the center with a bird on their shoulder, a top hat, a left red shoe, and a right blue shoe, some prompt information will likely be lost or elements misplaced. Blendbox’s real power lies in enabling control over an infinite array of elements, allowing you to position and size them precisely as desired, giving you mastery over the composition. This refined version improves clarity, flow, and readability while keeping the original meaning intact. Let me know if there’s anything else you’d like to adjust!” I invite you to read the full article here: https://lnkd.in/g_zXu6mw

16

1 Comment

The next few years in #AI are going to be wild. Check out this guitaring panda made by the new text-to-video model from China, KLING, that's rivaling OpenAI’s unreleased Sora model. KLING can generate up to 2-minute long videos with 1080p quality with just a text prompt 😲

7

Sharing insights on ever growing gaming industries both on rev and users base. How can we actually get these massive group of gamers to merge into Web3 ecosystem effectively. GAMES WORLD ASSET is the way. Let gamers have a chance to be a stakeholder in big IP games owning the rights.

15

Roblox is a platform to advertise on but most people don't realize how many unique ways there are to bring your message to life. There's programatic ads, custom integrations, owned bespoke experiences and even ads inside social media game/experiences like Clip It. Check out this article from Digiday written by Alexander Lee. Good insight from Ben Sarraille. I was happy to get my two cents in as well for Sawhorse Productions. It's exciting times for immersive marketing. If you aren't exploring these UGC gaming platforms, you aren't reaching Gen Z and Gen A as well as you might think... #roblox #ugcgaming #marketing #metaversemarketing #innovation

46

1 Comment

Proud of the team's work to help make this Genre update from Roblox possible! https://lnkd.in/gT5puNYF This is a truly unique platform, with some key takeaways for Roblox creators: 1) There are high-potential emerging genres for creators grow new audiences and drive innovation on the platform. 2) Four genres Roblox users are seeking are: open-world action, sports simulations, driving/racing, and social co-opetition. 3) It's powerful to localize content, build around communities, and create stream-worthy experiences. Interested in hearing thoughts from the community and game industry!

54

5 Comments

I've been following the newsletter for Mystery AI Hype Theater 3000. It's a fantastic, understandable source of realism in this environment of hype coming from big tech CEOs and pitch-people. Their latest issue touches on the residue left behind once the AI-bubble pops, and it is refreshing to hear folks think about the blowback of this cycle. Real people with real skills have lost jobs due to the _potential_ of AI — 10,000 layoffs in 2023 alone. These jobs are unlikely to come back. Data centers are expanding more quickly than the electricity to power them can support. Climate goals are being tossed aside. What happens to the infrastructure left over? Certainly it will continue to be used for data projects, putting any climate gains at risk. Its a slow-motion train wreck with real world consequences that very few are talking about. Highly recommend. https://lnkd.in/dTtt6Ueh

3

Black Mirror - AI Friend Wearable Spent $1.8m of $2.5m VC raised on one single domain name friend.com The sheer audacity and insanity of this. For gadgets selling for $99. The epitome of an outrageous publicity stunt. Or money laundering. This makes me mad. $1.8m could have bought education and food for entire villages in developing countries and you'd have made real friends. I wonder if the VC is kicking himself. This company has just screwed the entire industry. Now VCs are going to have to define allowable spending limits by line items and this is going to be painful for future startups raising funds. https://lnkd.in/e-FBNyDH

12

2 Comments

Exciting updates coming soon to our VPS SDK, including support for third-party scans, efficient map merging, and the ability to handle large areas for instant localization.

17

Thought this was clever https://lnkd.in/gmpPKzKR

2

This isn't just smart marketing - it's a testament to the power of pop culture and IP's in gaming. By immersing players in the world of "Love is Blind," the game doesn't just entertain; it creates a connection that extends beyond the typical 8-week season. #Gaming #MobileGaming #Innovation #intellectualproperty

5

Nice & easy to read high level explainer on identity signals.

6

I just watched this incredible demonstration of Gemini 2.0's native image output, and I’m completely blown away 🤯 by its capabilities. This model isn’t just another step forward — it’s a leap ahead in contextual Generative AI tech. 🤖 Unlike most other image-generation tools, this one stands out by keeping the subject consistent across prompts — a feature that feels like a total game-changer for anyone in creative fields, marketing, or even just experimenting with AI. 💡 What impressed me the most? The presenter asked the model to visualize a flying car soaring through the clouds — and not only did Gemini 2.0 keep the car consistent, but it also crafted an entirely (un)believable new scene, all in a single response! ✨ Imagine the possibilities: ✅ Creating cohesive brand assets. ✅ Generating storyboards with consistent character design. ✅ Enhancing your storytelling without losing focus. This level of reasoning, precision, and interactivity makes Gemini 2.0 a superior model compared to most tools out there. Whether you’re a designer, developer, or someone passionate about innovation, the potential here is truly next-level. 🚀 These new output modalities are only available to early testers (sadly, not me! 😅), with a wider rollout expected next year. Want a GDG Lahore session on it? 😉 🎥 Watch the video (in the comments 👇🏼) and let me know your thoughts: How do you see this shaping the way we interact with AI in creative projects? #AI #Gemini #GenerativeAI #Design #Creativity

20

4 Comments

JPEG image and MPEG video drives the daily infrastructure of business from streaming international conference calls to advertising product images over the internet but have you ever wondered how it actually works underneath the hood? The ideas are so advanced it's almost alien technology. Read on to find out how it actually works.

1

Find out more about what we did to compare DASH with MOQT.

18

Recently, I was part of a panel on generative AI at the TechCrunch Disrupt conference, moderated by Amanda Silberling. Amanda asked a great question, “But is it Art? How to harness generative AI in music and video production without displacing artists.” At Splice we support producers at various stages of their creative journey, from Grammy-winning artists to bedroom producers. With my experience building creator tools at GitHub and Adobe and most recently Splice, I think about this question all the time. First – there will be plenty of bad content made with AI – click bait designed to capture micro-seconds of attention in an increasingly noisy world, and in some cases, this will be designed to do real harm. So, yes – there will be “AI slop” but there will also be great things created with AI - and some of it will be art. And, the way we think of art will change over time. “Appalachian Spring" by Aaron Copeland won the Pulitzer Prize for Music in 1945, and Kendrick Lamar’s “DAMN” won the same prize in 2017.  Storytelling evolves, Art evolves, and with AI – it will evolve even faster. As more people have access to these tools, artists will take even greater risks and break boundaries - innovation will flourish, and art will move in extraordinary new directions. But before it can do that, as an industry, we MUST agree to do three critical things.  1. Respect the rights of creators. Provenance matters. We must ensure AI-assisted works don’t exploit the original contributions of artists. 2. Copyright clarity should be provided to everyone using these tools. Creators need to understand their rights around ownership of new creative works made with AI. 3. Build tools that enhance the creative experience. Preserving the integrity of the creative process is paramount, and technologists must collaborate with artists to build the future. These three things are fundamental to the next phase of evolution as AI continues to converge with creation. I agree, AI will be at its best when people don't know they’re using it - but I know people who love making music are not looking for AI to take away their creative experience.

117

7 Comments