Nuri Cankaya

Decisions made by AI should be transparent and traceable and the solution for that is using "Blockchain" for AI; here is the quick steps to create that: - Step 1: Develop a blockchain system that stores all the decisions made by AI models; blockchain data storage is immutable; similar to ledgers we use today; we cannot delete but add a line to the data. - Step 2: Build an interface that allows you to analyze AI outputs regularly with explainable AI practices. - Step 3: Whenever a bias-AI decision making process detected go deeper with the training dataset and fix the problem at the root and dig for more decisions made previously using the same wrong data input. We have to build a traceable AI with blockchain to mitigate AI mistrust

If want to measure the impact of your AI work effectively you should include sustainable AI & ethical AI practices you achieved with your AI project: - Are you using Sustainable AI: AI consumes lots of energy for compute and ensuring your AI project is scoped to use the right resources which are environmentally responsible, ethically designed, and socially beneficial is important. Do you need to use a cloud solution or you can use a local datacente or an AIPC to solve the need? Always consider the best option for sustainability. Foster long-term positive impact without compromising on your project outcomes. - Are you using Ethical AI practices: Prioritize fairness, transparency, accountability, and respect for user privacy and rights.

If you AI solution is not efficient or cost effective I recommend conducting a detailed cost analysis and find new alternatives to your AI solution: - Consider Leveraging Open Source: Rather than proprietary solutions like Cuda you can use open source models to easy develop and deploy your solutions. - Analyse Cost/Performance Analysis via MLcommons: MLcommons website will provide you detailed MLperf results from inferencing to training where you can find alternative solutions more cost effective and better performing. - Cloud/Hybrid/On-Premise: You should be able to choose where you want to run AI solution in this ever-changing AI landscape. Making decisions like Hybrid AI can boost your cost advantage and can bring security/privacy also

For business leaders understanding the benefits of AI will be a key competency to make a big difference in business outcomes; below are three resources I recommend: - AI Business School by Microsoft is a free education content by Microsoft Learn - Implications for Business Strategy by MIT is paid but couple of my team members took the course and great content. Worth the investment. - Finally I personally took this course from Harvard which seems technical, but I leveraged many of the learnings at business as well: Fundamentals of Neuroscience the Brain; free course. Great resource for leaders on AI journey.

Joining forces of human creativity and AI will bring unbelievable groundbreaking; Let's unpack with some examples: - Bio-Mechanics: Merging biotechnology & mechanical engineering requires close work between humans and AI to make this breakthrough. To expand human species life span and adapting rapidly with AGI will require bio-mechanics engineering that will use both human creativity but also deep learning approach of AI. -Hybrid Art: We already see some AI infused art today but human creativity will bring art to a new level with right AI LLMs that are bringing new dimensions to art. A good example is human retinal processing only understands certain light waves; where AI vision has unlimited potential to create new forms of hybrid art.

AI has been an area of research starting with defining "Intelligence"; I have found cross-disciplinary approach especially with neuroscience important for preparing for the future. As an example understanding how vision works with our own eyes with the visible spectrum and how AI can expand our ability to see beyond the light spectrum visible to us. Similarly how brain works and information is stored in neurons with a low electric energy in human brains today can influence much of the needed insights on todays energy hungry accelerated computing world. So many cross-discipline insights can be applied to bring AI to next stage for AGI.