Key Takeaways:

DeltaGen launches an AI platform to simplify generative AI use for companies, especially those without AI expertise.

Luma AI's Dream Machine faces IP infringement claims over characters similar to Pixar's "Monsters, Inc."

Meta introduces five new open-source AI models, including Chameleon, JASCO, and AudioSeal.


In today's episode of The AI Briefing, AI host Mick presents the latest in AI advancements. We start with Seattle-based startup DeltaGen, which is revolutionizing generative AI for enterprises. DeltaGen's platform, created by Rene Bystron, scans LinkedIn profiles to customize AI solutions for various job roles, eliminating the need for complex prompting. The non-chatbot interface selects from 30 AI models to deliver optimal outputs for different tasks. Recently accepted into the Techstars Workforce Development Accelerator, DeltaGen aims to target financial professionals with its innovative SaaS product.
Next, we discuss the controversy surrounding Luma AI's Dream Machine. The video generation tool is under scrutiny after producing a trailer for "Monster Camp" with characters resembling Pixar's "Monsters, Inc." This has sparked debates about IP infringement and the ethical use of AI, particularly regarding the datasets used in training. Despite inquiries, Luma has not clarified its training process, and Disney and Pixar have not yet commented on the situation.
Lastly, Meta has released five groundbreaking AI models. These include Chameleon, a mixed-modal model for generating images and text; JASCO, a text-to-music model; and AudioSeal, for detecting AI-generated speech. Additionally, new multi-token prediction models and geographic diversity evaluation tools aim to enhance language model training and text-to-image AI outputs, respectively. By open-sourcing these models, Meta encourages responsible innovation in AI technology.
ℹ️ The AI Briefing is an AI-generated podcast we created as an experiment to uncover what is and is not possible for automating the podcast production process. Information disclosed in the podcast episode is hand-picked and reviewed by humans prior to the episodes being created. We appreciate any feedback you have and appreciate you tuning in.

Sources:
https://www.databricks.com/company/newsroom/press-releases/databricks-introduces-new-generative-ai-tools-investing-lakehouse
https://www.geekwire.com/2024/ai-without-prompting-this-startup-just-joined-techstars-to-make-generative-ai-easier-to-use/
https://www.techtimes.com/articles/305824/20240619/luma-ai-video-tool-dream-machine-under-scrutiny-altered-version-disney.htm
https://www.theverge.com/2024/6/18/24181375/luma-ai-monster-camp-monsters-inc-pixar
https://www.pymnts.com/artificial-intelligence-2/2024/meta-releases-ai-models-that-generate-both-text-and-images/
https://ai.meta.com/blog/meta-fair-research-new-releases/