- AI Odyssey
- Posts
- Biden wants to ban AI voice impersonation...
Biden wants to ban AI voice impersonation...
Plus: Sam Altman returns to OpenAI's board
Hey everyone, welcome back to another edition of AI Odyssey!
Our insights this week are like a perfectly curated playlist – each one hits just the right note to keep you informed and engaged.
A LOT has been happening over the last week!
So let’s dive in!
Nvidia and HP join forces to accelerate generative AI workflows
In an exciting announcement, NVIDIA and HP have revealed that NVIDIA's powerful CUDA-X data processing libraries will be integrated into HP's AI workstation solutions.
No coding changes are required!
Specifically, the NVIDIA RAPIDS cuDF library will turbocharge the beloved pandas software used by millions of data scientists, delivering up to a staggering 110x speedup when paired with an NVIDIA RTX 6000 Ada Generation GPU compared to CPU-only systems.
Why it matters:
With datasets continuing to balloon in size, the integration of NVIDIA's GPU-accelerated software into HP's beastly workstations like the Z8 Fury means data scientists can now crunch through even the most massive generative AI workloads locally. RAPIDS cuDF for accelerated pandas is expected to land on HP AI workstations and PCs this month, with broader availability on HP AI Studio later in 2024.
Sponsored by HubSpot
Transform your workday with HubSpot's free guide, "Using ChatGPT at Work." With 100 actionable examples, learn how to navigate daily challenges, streamline processes, and revolutionize your workflow. Whether it's composing emails, generating content, or analyzing data, this guide is your key to leveraging ChatGPT for unparalleled efficiency.
Inside Your Guide:
A Treasure Trove of Prompts: A handpicked collection of 100 prompts that unlock the creative and problem-solving potential of ChatGPT.
Practical Insights: Learn how these prompts can streamline your daily tasks, from email automation to data analysis.
Expert Best Practices: Gain valuable insights into integrating ChatGPT effectively into your workflow for optimal results.
Step-by-Step Applications: From novice to expert users, these prompts are tailored to help you navigate through various business scenarios with ease.
Sam Altman returns to OpenAI's board
Remember that whole firing fiasco back in November? Yeah, about that…
The dust has now settled, and Sam Altman is returning to the board alongside some impressive new faces.
Why the change?
An investigation cleared him of any wrongdoing, revealing the cause to be a simple "relationship breakdown" with the previous board.
Altman will be joined by the former CEO of the Gates Foundation and the CEO of Instacart.
New governance rules:
To prevent future misunderstandings, OpenAI is revamping its governance with new rules and a whistleblower hotline. They're also tightening their conflict-of-interest policy, which is always a good idea for a transparent and fair policy.
Looks like everyone seems to be on the same page now! Is it the happily-ever-after for OpenAI and Sam Altman? 👀
Is Nvidia using published works to train its new AI? Sued by 3 authors
Another day, another lawsuit.
Nvidia has been sued by three authors - Brian Keene, Abdi Nazemian, and Stewart O'Nan.
The claim:
The authors claim that Nvidia used their copyrighted books without permission to train its NeMo AI platform for generating written language.
Specifically, the authors say their works were part of a dataset of around 196,640 books used to train NeMo before being taken down in October 2023 due to reported copyright infringement.
The lawsuit, filed as a proposed class action, alleges that the takedown reflects Nvidia admitting to training NeMo on the copyrighted dataset, thereby infringing the authors' copyrights.
Despite the lawsuit, Nvidia's stock soared 16.4% to a record $785.38 on the day the news broke, giving it a nearly $2.2 trillion market value amidst AI's boom. 😳
Hopefully, it serves as a warning and a lesson for all AI companies to be wary of the database they use to train their AI on.
💣Latest AI hits you can’t miss:
Amazon announced a $53M investment into Glacier, a female-led AI startup working to improve recycling efficiency and reduce waste.
Google unveiled new changes to its search ranking algorithm to combat the rise of AI spam and reduce low-quality content in results by up to 40%.
Ebay is building its own LLM trained on its e-commerce data, with plans to double its GPU capacity over the next year.
Microsoft is blocking AI prompts that led to Copilot’s Designer AI image creator generating violent and sexually inappropriate photos.
90 scientists signed a letter to prevent AI bioweapons like viruses or toxins, while still allowing beneficial applications.
Bloomberg found OpenAI's GPT-3.5 AI exhibited racial bias when ranking job candidate resumes, consistently scoring those with Black-sounding names lower than equally qualified applicants.
Biden wants to “ban AI voice impersonation.”. This comes after misleading robocalls targeted New Hampshire voters, hinting at broader restrictions beyond just political calls.
A Google engineer has been indicted for allegedly stealing over 500 confidential files containing trade secrets related to Google's AI while working for Chinese AI companies.
AI art to sip your coffee with…
Prompt:
**closeup of a blue/yellow/green/red ribbon made by melted glass, colorful murano, over a wooden table, christmas lights background
🕑 Latest AI tools:
"Ask Redfin" can answer questions about listed properties and neighborhoods, allowing users to inquire about details like amenities, pricing, and safety before connecting with an agent. Initially available in 13 U.S. cities.
Inscribe, this platform works to detect fraud in areas like business underwriting, tenant screening, and onboarding.
Brevian is a no-code enterprise platform for building AI agents.
Midjourney’s v6 turbo mode is out. Make images at 3.5x the speed (for 2x the cost), claims to be one the fastest GPUs in the world.
How ‘Ask Redfin’ chatbot looks like.
That’s all for today! Thanks for tuning in, see you on Thursday! 🤗