Victor Tanashchuk/Getty Images
The artificial intelligence (AI) market is currently worth approximately $9 trillion, and businesses are just beginning to explore its potential. AI chip maker Nvidia and model maker OpenAI will take the lead and ultimately compete against each other in the AI race.
The latest State of AI report analysis, co-authored by AI investor Nathan Benaich and Alex Chalmers, head of platforms at Air Street Capital, provides insight into the current state of AI and the potential for the future of emerging technologies. The observations and predictions are explained in detail.
Also: Salesforce Research Says CIOs Should Also Be Chief AI Officers
Their predictions for AI in the coming year include:
“Apps and websites created only by people with no coding ability go viral (e.g. App Store Top 100).” “Open source alternatives to OpenAI o1 outperform it across a variety of inference benchmarks.” “Challengers have meaningful impact on NVIDIA’s market position.” “Apple’s impressive results from on-device research accelerate momentum for personal on-device AI.”
Mr. Benaich’s predictions from past reports have largely come true. Among a few that missed the mark was a prediction that “the Gen AI scaling craze will see groups spend more than $1 billion training a single large-scale model.” They admit this hasn’t happened yet, but they’re willing to “give it another year.”
And do you think AI can solve all your business problems? Apple’s new study shows otherwise
They also predicted that “major AI companies (such as OpenAI) will acquire or build inference-focused AI chip companies as inference workloads and costs grow significantly.” They said, “Sam Altman is reportedly raising huge sums of money to do this, while Google, Amazon, Meta, and Microsoft each continue to build and improve their own AI silicon.” points out.
The co-authors also investigated the state of AI implementation within companies. “Enterprise automation is set for an AI-first upgrade,” they observed. This includes AI-powered robotic process automation (RPA).
Related article: OpenAI’s Altman sees ‘superintelligence’ just around the corner, but not enough details
“Traditional robotic process automation, such as UiPath, suffers from high setup costs, unstable execution, and cumbersome maintenance,” they said. “Two new approaches, FlowMind (JP Morgan) and ECLAIR (Stanford), use underlying models to address these limitations.”
For example, he explained, “FlowMind achieved 99.5% accuracy in workflow understanding.” At the same time, “ECLAIR takes a broader approach, using multimodal models to learn from demonstrations and interact directly with graphical user interfaces across a variety of enterprise settings. For web navigation tasks, ECLAIR We increased our completion rate from 0% to 40%.”
Benaic and Chalmers have focused a lot of attention on AI chip producer Nvidia, observing that the company “remains the most powerful company in the world, enjoying its place in the $3 trillion club.” did. “NVIDIA has already booked significant pre-orders for our new Blackwell GPU family, has a strong strategy in place for governments, and our ambitions are only growing.”
Another element of the AI world, OpenAI, also received a lot of space in the report. “OpenAI’s reign of terror is over until Strawberry comes along. [o1] Citing OpenAI’s latest model, Benaich and Chalmers said it has “landed a doubling of the scaling of inference computation. Even after the launch of o1, OpenAI continues to We maintain our dominance.”
Related article: Welcome to the AI revolution
By moving the AI’s computations from pre- and post-training to inference, o1 can incrementally reason through complex prompts in a Chain of Thought (COT) style and employ RL to enhance the COT and the strategies it uses. ” they added. “This opens up the potential for LLMs to solve multi-layered math, science, and coding problems that have traditionally struggled due to the inherent limitations in predicting the next token.”
The AI market is huge at the moment. The authors estimate the value of AI companies at $9 trillion, but say “the level of investment in private companies is growing at a healthy rate.” Startups are “starting to gain traction in areas such as video and audio generation,” he added.
