Move over to Generative AI. Agentic AI and Vertical LLMs are the hottest things in the market. There are millions and billions of dollars in opportunities out there. Little did I realize as a kid when I grew up that Rags to Riches had a different meaning. The new word is RAG TO RICHES.
Amid the excitement about the possibility of artificial intelligence, the measured voice of Massachusetts Institute of Technology professor Daron Acemoglu, PhD, offers a sobering perspective. Dr. Acemoglu calculates a modest 5% of jobs to be replaced or heavily supplemented by AI over the next decade. 60% of Enterprises are unable to Justify the costs or get ROI.
Coming back from
ISTARI
event in collaboration with the University of Cambridge, Global CISOs barely talked about basic AI as people are getting fatigued with the hype.
Concerns about the aging of AI, Increasing tech debt, and security vulnerabilities arising, Let's discuss the perils of Ignoring Tech checks for AI
- The Grip of Obsolescence: Technology evolves at a breakneck pace. We now have Right's law and with QUANTUM, we will have new laws. What was considered groundbreaking a few years ago can quickly become outdated. Libraries and frameworks can become unsupported, hindering performance and leaving your AI system vulnerable to security exploits. Imagine a system relying on a facial recognition library no longer maintained by its developers. This could lead to the system failing to recognize faces accurately or even becoming susceptible to spoofing attacks. People who understand the #SBOM have a nightmare with existing software and are now forced to grapple with newer challenges where models are coming in days and weeks.
- Compatibility Chaos: As the tech landscape changes, new systems and tools emerge. Without regular assessments, your AI system might become incompatible with these advancements, creating hurdles in data exchange and integration. Think of an AI-powered customer service chatbot built using a framework no longer compatible with the company's new communication platform. This would prevent the chatbot from functioning effectively and leave a negative customer experience. Of course, we have agenticai to build connectors but that is also tech debt.
- Missing Out on the Future: The field of AI is constantly innovating. New business models are expected to be created. It is expected that AI is likely to create 97 million jobs and add 1Trillion $ to U.S.? GDP by 2030. Sadly we are losing jobs at a crazy pace and I am not sure if it is net new jobs or what. New algorithms, processing techniques, and hardware capabilities are unveiled at an accelerated pace. By failing to assess your system's tech stack, you might miss out on opportunities to improve its capabilities and performance. For instance, an AI system for medical diagnosis might not be able to leverage the latest advancements in natural language processing, hindering its ability to analyze medical reports and patient data effectively.
- we created a digital divide and have caused all kinds of new disorders and are using tech yet again to address the empathy factor. As we become increasingly dependent or smart, intelligent friends and tech obsolence will play a critical factor
- Enterprise Observability:- Sadly everyone has their definition and some are not even talking about system reliability at all. We are not talking of LLMops and the standard hype-cycle of multi-billion dollar ARR with it and just a few months back it was AIops and before that, it was MLops. oops, I am taking off too many ops. At the end we need Devops but now we are Bizops, Processops, Dataops, SRE, etc. Are we too obsessed with AI that basic things of #Businesscontinuity hardly matter? #CROWDSTRIKE is a classic example. #
As I continue to quote from
Seth Earley
" There is no AI without IA ( Information Architecture= Business+Data+Applications+Technology+Security+Sustaianability)"
With all the challenges said, AI will continue to evolve but we need to separate hype from reality and the future strongly lies in our hands think IA to have a secure, sustainable, and scalable system. ( Courtesy:- AWS Reference Architecture)
The Benefits of Regular Tech Assessments for AI Systems
Regularly assessing the technology used in your AI systems offers a multitude of benefits that ensure their long-term viability and effectiveness. Here are some key advantages:
- Combating Obsolescence: By proactively identifying outdated technologies, you can take steps to migrate your AI system to support alternatives. This ensures your system remains secure, and efficient, and can leverage the latest advancements offered by the chosen libraries and frameworks.
- Maintaining Compatibility: Regular assessments help identify potential compatibility issues with newer technologies. This allows for proactive planning and implementation of solutions to ensure your AI system can seamlessly integrate with evolving tools and platforms.
- Future-Proofing Your System: The landscape of AI is constantly in flux. Tech assessments help you stay informed about emerging trends and advancements. This knowledge empowers you to identify opportunities to improve your AI system's capabilities and ensure it remains competitive in the long run. Imagine an AI system for predicting stock market trends. Regular assessments can help identify new machine learning algorithms or data sources that could improve the system's accuracy.
- Enhanced Security: Outdated technologies often harbor security vulnerabilities. Assessments can help identify these vulnerabilities and allow for timely patching or migration to secure alternatives. This safeguards your AI system from cyberattacks and data breaches.
- Improved Performance: Newer technologies often offer better performance and efficiency. Assessments can help identify opportunities to upgrade your AI system's tech stack, leading to faster processing times and improved overall performance.
By incorporating an IA mindset, Enterprises can ensure systems continue to operate at peak efficiency, remain compatible with evolving technologies, and are positioned to benefit from the latest advancements in the field. Yes, we can also hire #Chiefcleanupofficers too.
Chief Revenue Office at Intangles Lab
1 个月Insightful