insideBIGDATA AI News Briefs – 11/22/2023

Print Friendly, PDF & Email

Welcome insideBIGDATA AI News Briefs, our timely new feature bringing you the latest industry insights and perspectives surrounding the field of AI including deep learning, large language models, generative AI, and transformers. We’re working tirelessly to dig up the most timely and curious tidbits underlying the day’s most popular technologies. We know this field is advancing rapidly and we want to bring you a regular resource to keep you informed and state-of-the-art. Enjoy!

OpenAI in Chaos

The upheaval at industry highflyer OpenAI started last Friday (Nov. 17, 2023) when the company’s board of directors fired CEO and co-founder Sam Altman and appointed Mira Murati, the CTO, as acting CEO. Following this event that caught the industry by surprise, employees and investors advocated for Altman’s reinstatement. However, the board approached Nat Friedman, a former GitHub executive, to take over the CEO position, however he declined the offer. Shortly thereafter, Emmett Shear, the former CEO of Twitch, agreed to take on the role. Early on Monday, before the opening of the stock market, Microsoft announced the hiring of Altman and several other individuals to spearhead a new venture dedicated to AI technology. This was NOT the end of the drama!

“Mr. Altman’s departure follows a deliberative review process by the board, which concluded that he was not consistently candid in his communications with the board, hindering its ability to exercise its responsibilities,” the company said. “The board no longer has confidence in his ability to continue leading OpenAI.”

By Monday afternoon, around 710 of OpenAI’s 770 employees revolted and signed a letter urging the resignation of the company’s board members, and demanding the reinstatement of Sam Altman and former President Greg Brockman. The letter indicated that if Altman and Brockman are not reinstated, the employees are prepared to follow them to Microsoft. Board member and chief scientist Ilya Sutskever’s expressed regret over participating in Altman’s firing. Microsoft stock hit an all-time high on Monday after announcing the hiring of Altman, gaining over $115 billion in value compared to Friday after market low.

Continuing to move forward on Monday, a majority of the 24 key leaders at the company had either signed the letter or resigned. The letter criticizes the board for undermining OpenAI’s mission and competence in overseeing AI technology development. The board refuses to share the circumstances leading to Altman’s firing. Rumor has it that there was strain between Altman and OpenAI’s board over issues of AI safety, the pace of technology development, and the company’s commercialization strategy.

It is believed that Altman’s side ventures contributed to the his falling out with the board. He was fundraising in the Middle East for a new chip venture to rival NVIDIA before OpenAI’s board ousted him. He was also raising funds for an AI-powered hardware device, collaborating with former Apple Inc. design chief Jony Ive.

Moving on to Tuesday, rumor has it that Marc Benioff of Salesforce has indicated that his company will match any OpenAI researcher who has tendered their resignation full cash & equity OTE to immediately join the Salesforce Einstein Trusted AI research team under Silvio Savarese. Benioff is soliciting CVs now. The same offer was extended by Kevin Scott, CTO of Microsoft to join Altman at Microsoft’s new AI research lab. And it appears that the OpenAI board has reached out to competitor Anthropic to merge the companies for an increased capability vector. Per Steve Sloan of Menlo Ventures (investors in Anthropic), all the signals seem to indicate that Altman will be back at OpenAI by Wednesday.

“I believe startups that have been coasting as a wrapper and/or relying on OpenAI with a sophisticated integration could be at risk or at the mercy of Microsoft if the company goes down…potentially creating a world where only those with advanced AI teams will prevail by building private models, said Sarah Nagy, CEO & co-founder of Seek AI.

Speculation in Silicon Valley suggests that the board apparently felt Altman was moving too quickly, with too much focus on rapidly deploying consumer products contrary to OpenAI’s non-profit mission of “AGI that benefits of humanity” and widespread concern over safety. There may have been conflict behind Sam’s attempts to achieve AGI. Altman stated to Congress, that he takes existential risk from AI seriously, but his actions may speak louder than words, with government actors perhaps not buying it, leading to political pressure on the board to rein him in.

Coming full circle, late Tuesday evening OpenAI issued a Tweet indicating that the company reached an agreement in principle for Sam Altman to return to OpenAI as CEO with a new initial board of Bret Taylor (Chair), Larry Summers, and Adam D’Angelo. Whew! What a whirlwind.

Kyutai Raises $330 Million to Open Source Everything

Paris-based Kyutai is a non-profit laboratory entirely dedicated to open research in AI. Its objective is to tackle the main challenges of modern AI, particularly by developing large multimodal models – using text but also sound, images, etc. – and by inventing new algorithms to enhance their capacities, reliability and efficiency. To do this, the laboratory will use the computing power made available to it by Scaleway, an iliad Group subsidiary. Scaleway’s supercomputer has the highest-performance computing power for
AI applications deployed to date in Europe. Resolutely committed to the democratization of
AI, Kyutai is positioning itself as a leading player in AI open science. Its ambition is to share its advances with the entire AI ecosystem – the scientific community, developers, companies, society at
large and decision-makers in democracies. Kyutai will also contribute to the training of future AI experts, by offering internships to students on Master’s programs and supervising PhD students and postdocs.

Microsoft Adds Generative AI Models to Azure Model Catalog

Microsoft introduced new models to the Azure AI model catalog, including Nemotron-3 8B, Code Llama, and Mistral. They also introduced ‘Models as a Service’ (MaaS) for easier AI model integration and customization.

NVIDIA Introduced Nemotron-3 8B to Revolutionize Enterprise AI Development

NVIDIA introduced the Nemotron-3 8B family, a set of generative foundation models within its NeMo framework. This family of models is designed to improve enterprise AI applications, and includes base, chat, and question-and-answer checkpoints. The Nemotron-3 8B models are available on the Azure AI Model Catalog, HuggingFace, and the NVIDIA AI Foundation Model hub, and can be fine-tuned for custom use cases.

Nvidia’s NeMo framework enables enterprises to quickly implement AI applications across various environments. The Nemotron-3 8B family gives developers an easy way to integrate foundation models and fine-tuned versions of these models into enterprise-specific requirements.

Trillion Parameter Consortium launches with dozens of founding partners from around the world 

A global consortium of scientists from federal laboratories, research institutes, academia, and industry has formed to address the challenges of building large-scale artificial intelligence (AI) systems and advancing trustworthy and reliable AI for scientific discovery.  

The Trillion Parameter Consortium (TPC) brings together teams of researchers engaged in creating large-scale generative AI models to address key challenges in advancing AI for science. These challenges include developing scalable model architectures and training strategies, organizing, and curating scientific data for training models; optimizing AI libraries for current and future exascale computing platforms; and developing deep evaluation platforms to assess progress on scientific task learning and reliability and trust. 

TPC represents a practical approach to surmount existing limitations in AI model training and data processing. Its emphasis on optimizing AI libraries for exascale computing and developing effective evaluation methodologies addresses some of the key technical challenges in advancing AI applications in scientific research. 

SambaNova Comments On Chips / Disruptors On NVIDIA’s Heels

With Q3 earnings and the upcoming launch of NVIDIA’s H200 chip (which won’t be available until next summer), SambaNova’s CEO and Co-Founder, Rodrigo Liang offers the following commentary:  

“NVIDIA continues to perform well and the demand for AI is clearly rocketing, but end users want choice and there are suddenly more options for training and running Generative AI models. Microsoft just announced two alternatives to NVIDIA’s chips for its Azure cloud, for example.

As AI workloads become more pervasive, larger and complex, new chips, models and solutions are emerging that enable customers to run workloads more efficiently. The market indicates that people want more innovation to choose from, so we’ll start to see lanes emerge within the industry, where competitors with disruptive technologies offer customers more choice or specialization. 

For a technology that NVIDIA CEO Jensen Huang himself said is going to be bigger than the internet, it’s the innovators that will come out top, and there are plenty of competitors nipping at NVIDIA’s heels.” 

In addition, in light of the looming chip shortage, NVIDIA is not the only chip in town. SambaNova Systems, makers of the purpose-built, full-stack AI platform, recently announced its newest generation chip: the SN40L. This ‘truly intelligent’ chip includes both dense and sparse compute, and both large and fast memory. Specifics around SN40L include: 

  • Serving a 5 trillion parameter model, with 256k+ sequence length possible on a single system node. This enables higher quality models, with faster inference and training at a lower total cost of ownership.
  • Larger memory unlocks true multimodal capabilities from LLMs, enabling companies to easily search, analyze, and generate data in these modalities.
  • Lower total cost of ownership (TCO) for AI models due to greater efficiency in running LLM inference.

AI Ready

Amazon plans to offer free ‘AI Ready‘ courses to teach AI Skills to 2 million people by 2025 as competition with Microsoft heats up.

Sign up for the free insideBIGDATA newsletter.

Join us on Twitter:

Join us on LinkedIn:

Join us on Facebook:

Speak Your Mind