Turning OpenAI Back on a Twist: Timing the Fate of the CEO and Breaking the Plug in A.I.

OpenAI was founded about eight years ago by Altman, Brockman, Musk and others. Under Altman’s leadership, it transformed into a lucrative juggernaut. The company released a product last year that set the pace for the tech industry’s focus on a more advanced form of artificial intelligence known as generative Ai.

The directors believed that they were doing their job when they fired Altman, and as a result they believed they were doing their job of making sure the company developed powerful artificial intelligence safely. Increasing profits or ChatGPT usage, maintaining workplace comity, and keeping Microsoft and other investors happy were not of their concern. The directors, including Adam D’Angelo, Helen Toner, and Sutskever were not happy with the wayAltman dealt with them. Bottom line: The board no longer trusted Altman to pursue OpenAI’s mission. If the board can’t trust the CEO, how can it protect or even monitor progress on the mission?

Two signal events have happened in A.I. in recent weeks. You’ve heard about that one. Sam Altman, the company’s chief executive, was fired by the nonprofit that governs OpenAI. The decision was sudden and left a lot to be desired. “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,” read a cryptic statement.

Artificial intelligence researchers and science fiction writers fear that the machine cannot be turned off. A powerful A.I. is developed in the story. Its designers are thrilled, then frightened. They pulled the plug, only to discover the A.I. had copied its code elsewhere.

OpenAI, Stable Video Diffusion, and Anthropic: Updates on Claude 2.1 and the Rate Limit for GPT-4 Turbo

OpenAI said late Tuesday that it had rehired Sam Altman as its chief executive, after a tumultuous five days that included acrimony between the company’s board and management.

The company, maker of the popular ChatGPT, said it would also create a new board of directors. This comes after the former board voted to fire Altman as CEO late last week.

Sutskever did not like the joke that the weird org chart that mapped out the relationship looked like a future GPT might come up with. He pointed out that we’re the only company in the world with a capped profit structure. If you believe, like we do, that if we succeed really well, then these GPUs are going to take my job and you and everyone’s jobs, it seems nice if that company would not make truly unlimited amounts of returns. The board is keeping an eye on things so the profit-seeking part of the company does not shirk its commitment to make sure the machine doesn’t get out of control.

The board that fired Sam Altman last Friday did not have any confidence in the CEO because he was not consistently candid in his communications with them. No examples of that alleged behavior were provided, and almost no one at the company knew about the firing until just before it was publicly announced. Microsoft CEO Satya Nadella and other investors got no advance notice. The four directors, who are a majority of the board, kicked Greg Brockman off the board. Brockman left quickly.

Stable Video Diffusion was released by Stability Artificial Intelligence and Anthropic announced a more powerful version of its Claude chatbot after the sudden departure of OpenAI’s leadership.

Anthropic’s latest model, Claude 2.1, was given two key updates. One is the ability to upload more data at once to the chatbot and fewer lies. The token limit for Claude is now set at 200,000 tokens, which is approximately the length of a 500 page book. (Sorry Leo Tolstoy fans, you’ll have to wait until future updates to analyze all of War and Peace in a single prompt.) To compare, the rate limit for the GPT-4 Turbo model, announced by Altman pre-firing, is capped at 128,000.

Anthropic claims that the new Claude is more likely to admit when it is unsure of an answer than it is to fib with great confidence. The company said in a post that they tested Claude 2.1’s honesty with a large set of complex, factual questions that probed known weaknesses in current models. A lack of veracity, often described as hallucinations, continues to be a major issue for chatbots.

Unveiling the Puzzles of Video and Text-to-Face: The Case of Stable Video Diffusion and Claude 2.1

When you input a prompt into a text-to-video model, the AI spits out GIF-like animations that can range from eerily beautiful to downright disturbing. In addition to text-to-video capabilities, Stable Video Diffusion can transform your still images into videos by adding motion.

While this isn’t technically a new feature from OpenAI, the company rolled out ChatGPT with voice capabilities to everyone in the short period while Altman was out as CEO. The feature used to only be available to those who paid for OpenAI’s $20-a-month subscription service.

It’s not yet giving Spike Jonze’s Her, but the software developers at OpenAI took another big step towards their goal of “multimodality” by giving the chatbot the ability to hold a conversation with you. The idea is that a chatbot can be even more powerful if it can accept inputs and provide outputs in multiple mediums, like voice, text, and images. Who knows when it’ll learn how to smell.

It feels like every week there’s a new launch or announcement from one of the major players. When I read about Stable Video Diffusion and Claude 2.1, I assumed they were a coincidence, says Dharmesh Shah, who was the CTO and co-owner of HubSpot at the time.