Artificial intelligence

Artificial intelligence (AI) is the theory and development of computer systems that can perform tasks that would otherwise require human intelligence, such as pattern recognition from data; learning from experience; interpreting visual inputs; and recognizing, translating, and transcribing language.

In practice, AI is a way for humans and computers to work better together. In design and make industries, AI helps creative people do more with less by automating mundane and repetitive tasks and by cutting through the complexity of huge datasets to provide insights that may lead to new solutions.

Artificial intelligence, or AI, refers to computer systems that can perform tasks that otherwise would require human intelligence. There are, however, numerous types of AI for different purposes. The following are some of the most important types of AI, all of which can help people and computer systems to work better together.

Machine learning (ML) is a subset of AI that lets computers “learn” from data sets without being explicitly programmed. Supervised learning ML learns from labeled data, such as images with metatags, whereas unsupervised learning ML works off unlabeled data. Either way, ML systems permeate modern life as a part of GPS maps, search and recommendation engines, spam filters, credit card fraud detection mechanisms, and much more.

Deep learning uses machine learning algorithms with biologically inspired logical structures called “artificial neural networks.” Deep learning is used for applications such as facial recognition and other types of computer vision, speech recognition for translating languages and transcribing spoken audio, bioinformatics such as analyzing DNA sequences and protein structures, and medical image analysis used to detect and diagnose diseases.

Generative AI is based on pre-trained, deep neural networks, which are capable of producing novel text, images, audio, video, and computer code. Especially when it comes to the large language models (LLMs) like ChatGPT, Claude.ai, and many others that produce written outputs, generative AI’s messages are probable but unvalidated.

Generative design, not to be confused with generative AI, is an algorithmic process in which software runs precise real-world simulations to arrive at a set of optimal design solutions for a defined problem. Designers or engineers define their design goals, along with constraints pertaining to cost, materials, manufacturing methods, and performance and spatial requirements. The software can quickly generate numerous options that explore the possible permutations of the design requirements, simulating and learning from each iteration. Generative design solutions often but do not always employ AI.

Ideally, AI will give users more time for creativity by decreasing the amount of tedious, repetitive chores. AI should augment and serve human creativity rather than replace it.

Generative AI has the potential to accelerate the visualization of concepts, leading to more ideas from more people. With faster ideation, teams can explore more possibilities and start more projects. Furthermore, AI features can help greatly reduce the gap between client expectations and design-and-make reality, boosting customer satisfaction. 

Generative AI is already showing its transformative power for creative software in the AEC, design and manufacturing, and media and entertainment industries. Architects will be able to generate editable floor plans and easily import information from drawings into CAD projects. Construction IQ’s data analysis in the Autodesk Construction Cloud identifies risks and informs decisions on design, quality, safety, and project control priorities. Product designers can generate part assemblies for 3D CAD models. And in media and entertainment, AI generates 3D character rigging, making skeletal animation much faster.