Artificial Intelligence (AI) is evolving quickly, and the impact is already widespread. Governments and organizations have taken notice and started to take on some of the challenges ahead. Policymakers must grapple with how to reap the potential benefits of AI while reducing potential harms. Governments have begun exploring approaches to AI.

AI technologies are moving quickly, and organizations must work hard to gain access to the best and most current AI knowledge. Generative AI has been of particular interest due to its ability to create brand-new output or content. This means new text, images, code, and renderings can be created based on the data on which the AI is trained. 

Organizations are working to assess generative AI and determine how these tools should be tested. For example, Singapore identified generative AI risks and set up a foundation, the AI Verify Foundation, to guide generative AI adoption. Among the key risks are hallucinations, accelerated disinformation, copyright issues, and embedded biases. The AI Verify Foundation proclaims to facilitate responsible AI adoption and promote best practices and standards. Members of the foundation include IBM, Zoom, Salesforce, and Hitachi.

The White House released an Executive Order on the safe development of AI at the end of October 2023. This Executive Order required developers to share safety test results and other information with the US government. The UK has also recognized the need to govern AI and held the UK AI Safety Summit in early November of 2023.

Some of the questions regulators must face include what types of work should AI be allowed to do. How much decision-making control should AI have? What safeguards should be in place? 

As argued by Open AI, regulators also need to face the path AI development will take in the coming decades. In the next ten years alone, it is possible that AI systems will exceed human expertise in many domains. Leveraging AI to create a more prosperous future for humanity requires managing risks along the way. Ideas for oversight include creating an international authority to inspect AI systems and placing restrictions on degrees of deployment and levels of security. Any such oversight should include citizen input and public oversight. AI governance should also foster creativity and growth by allowing open-source projects and companies to develop models that are below a capability threshold to avoid burdensome audits or licenses. 

AI, Architects and Design Professionals 

While AI is unlikely to replace architects and design professionals it has the potential to automate repetitive tasks. This could allow architects and design professionals to focus on the creative and strategic parts of design. AI could be used to optimize design alternatives, generate visual designs, cost estimation, analyze site data, recognize patterns in datasets, assist in software development, improve energy efficiency, analyze BIM data, and manage and review relevant data. Properly leveraged, AI could elevate the roles of architects and design professionals.  

Before reaching such heights, the foundation on which we build AI systems needs to be evaluated. This foundation is data. The promise of AI hinges on the data.

AI and Data

While many are excited by the potential end results and opportunities of AI, discussions of data management are less alluring and attractive. However, this foundational aspect is crucial to get right. The once popular idiom “Garbage in, garbage out” applies today to AI. Data is the fuel behind AI. The lower the quality of the data, the lower the quality of the AI solutions. 

Data is the beginning of all AI processes. Data management needs to ensure data is relevant, organized, and secure. The global data landscape is complex, vast, and siloed. Overhauling data by gathering scattered data will be required for creating effective AI solutions.

Data types are disparate and can include videos, images, and unstructured sources. AI can be leveraged to improve data management and can automate, clean, and organize datasets in preparation for advanced analytics.

As companies and organizations try to deal with their data, AI techniques can be leveraged to identify and correct errors in data. AI can facilitate data migration. AI can also be used to automate data integration from multiple sources, data tagging, data deduplication, data storage, retrieval, and compression optimization.

Data Management for Building Product Manufacturers: All in One Place 

One way building product manufacturers can prepare to take advantage of AI is to start by ensuring their data is clean and organized. As much as possible, data should be managed in a single location.

Building product manufacturers will want to have data on how the design community interacts with their products and design files. CADdetails’ Design Hub is a foundational platform for clients to collect data, and gain insight into architects' and design professionals’ interactions with their design content.

If you are a building product manufacturer interested in having your design content organized, managed, and generating valuable data, book a call with CADdetails today.

Cover image by Google DeepMind on Pexels