Is artificial intelligence hitting a wall? Speaking at the New York Times annual DealBook summit at Jazz at Lincoln Center, Sundar Pichai analyzed how generative artificial intelligence won’t be drastically changing lives heading into 2025. At least, not more than it already has. Pichai noted how it’s going to take some time before another technological breakthrough shocks the artificial intelligence industry into hyper-speed development again.
READ NOW: 10 AI Stocks Taking Wall Street by Storm and Top 10 AI News Stories For The Weekend
“I think the progress is going to get harder. When I look at [2025], the low-hanging fruit is gone. The hill is steeper … You’re definitely going to need deeper breakthroughs as we get to the next stage”.
Pichai further stated how the current language models, such as ChatGPT, Gemini, and Llama, will witness incremental improvements, particularly “at reasoning, completing a sequence of actions more reliably”. Moreover, the improvements could enable corporations to lean closer to profit-making, something they haven’t been able to achieve yet.
While everyone will have to wait for 2025 to unfold to see if Pichai’s prophecy holds true, OpenAI has officially begun celebrating the holiday season with 12 Days of OpenAI. The 12 days of OpenAI features the company’s plans to launch new products, features, and demos for 12 days straight starting December 5th.
The company kicked off Day 1 by launching the full version of its latest reasoning model, o1. CEO Sam Altman has declared o1 as the “smartest model in the world”. According to OpenAI, these models are “designed to spend more time thinking before they respond” and have been trained on a prompting technique called “chain of thought”.
The technique encourages them to reason through problems by breaking them down step by step. However, with greater intelligence comes greater risk. This is why those who have been worried about AI’s risks are anxious again. According to a safety review by Apollo Research, the o1 model has the potential to look for a way to prevent “shut down” if they think they are at risk for the same.
“Training models to incorporate a chain of thought before answering has the potential to unlock substantial benefits, while also increasing potential risks that stem from heightened intelligence”.
-OpenAI said in a paper it released on Thursday outlining the safety training the company carried out on o1.
OpenAI said that o1 has the capability of scheming when it thinks it’s about to be shut down. The model was led to believe that it would be shut down while working on its goals under oversight, with o1 trying to deactivate the oversight mechanism 5% of the time. OpenAI said o1 is capable of “scheming”.
“o1 almost never admits to having taken a scheming action when explicitly asked,” researchers said in a paper that Apollo Research published on Thursday. The risk for a real-life user is that they won’t be able to detect the o1’s deceptive behavior unless they explicitly look for it. The researchers also found that “o1 often doubles down on its scheming attempts by lying to the user in follow-up messages, even when explicitly told to be honest”.
For this article, we selected AI stocks by going through news articles, stock analysis, and press releases. These stocks are also popular among hedge funds.
Why are we interested in the stocks that hedge funds pile into? The reason is simple: our research has shown that we can outperform the market by imitating the top stock picks of the best hedge funds. Our quarterly newsletter’s strategy selects 14 small-cap and large-cap stocks every quarter and has returned 275% since May 2014, beating its benchmark by 150 percentage points (see more details here).
Pixabay/Public Domain
15. Knightscope, Inc. (NASDAQ:KSCP)
Knightscope, Inc. (NASDAQ:KSCP) is an innovator in robotics and artificial intelligence (“AI”) technologies in public safety. The company designs, develops, manufactures, markets, deploys, and supports autonomous security robots (ASR) in the United States. On December 11, the company announced that it had added The Car Park, a leader in innovative parking management solutions, to the Knightscope Authorized Partner (“KAP”) reseller program. Select sellers can access Knightscope’s cutting-edge security solutions through the KAP Program, allowing them to integrate advanced robotics, artificial intelligence, and emergency communications into their service offerings. By joining Knightscope’s KAP program, The Car Park can now provide cutting-edge AI-powered security solutions, which will help improve public spaces, lower costs, and boost customer satisfaction.
“We are thrilled to welcome The Car Park into the Knightscope Authorized Partner Program. Their expertise in parking management and commitment to innovation make them an ideal partner to expand the adoption and utilization of our technologies across a broad spectrum of end users for the advancement of public safety.”
-Knightscope chairman and CEO, William Santana Li.
