The term AI breakthrough is used to describe any significant technological advancement that demonstrates an artificial intelligence system’s ability to perform tasks or solve problems at a level beyond what humans can. The most prominent examples include the ability of autonomous AI agents to act without human direction and use machine learning algorithms for decision-making and interpreting data. This includes retail companies using AI to predict customer buying behavior and generate customized marketing copy and promotions, or logistics AI programs that can automatically adjust delivery schedules based on stock levels.
In the earliest decades of AI, researchers built robots that could follow specific rules or instructions to perform repetitive tasks in highly controlled environments. These robots, known as traditional AI, are still used today in manufacturing and other industries where consistency is key.
More recent AI breakthroughs have included the development of quantum computers. These machines operate at much higher speeds and scale than classical supercomputers by taking advantage of the properties of qubits, or bits of information stored in quantum states. One of the biggest AI breakthroughs this year was Google’s Willow, a 105-qubit chip that solved complex computational problems in five minutes — a task that would take classical supercomputers 10 septillion years to complete.
Another AI breakthrough has been the creation of generative models that can create content or images based on input or prompts, such as a text input, shapes in an image or frames of a video, or commands in software code. Variational autoencoders (VAEs), first seen in 2013, and generative adversarial networks (GNoME) are among the most popular and headline-making generative AI models. NVIDIA and Stanford recently demonstrated a new technique that can train an algorithm to generate Tom and Jerry-style cartoons with strong temporal and spatial cohesion.
About the Author
