CIO In The Know – AI Fatigue is coming and what’s next
Whew! Typically the summer months are slower than the rest of the year, but this summer has been nothing but crazy busy!
I attended plenty of great industry events this summer. In June, there was HPE Discover, Snowflake Summit and Databricks Summit. July was Splunk conf23 and then AWS’ GenAI Summit. August started with KBCM’s Technology Leadership Forum then Five9’s CX Summit and Google Cloud Next. And September had Salesforce’s Dreamforce and Workday Rising.
CIO Water Cooler Talk
Today, I spent the day with fellow CIOs at the Wall Street Journal’s CIO Network Summit in New York City. The agenda was all about AI…which is unusual to have an entire summit dedicated to a single topic. Why? The reoccurring topic du jour for CIOs continues to be AI. Generative AI to be specific. The conversations at the summit were incredible, but I will need a series of missives to unpack everything discussed and the impact for CIOs.
The good thing about the generative AI conversations is that a dose of rationalization is setting in. While initial conversations were abnormally slighted toward the upside and potential of generative AI, CIOs are now seeing more risk than opportunity. That’s not a rally call to stop experimenting with generative AI, but to tap the brakes and proceed more cautiously.
For some time, CIOs have considered how to bring guardrails into the conversation around generative AI as a means to protect sensitive and confidential data. Even while the major cloud service providers are delivering solutions to lower the hurdle for enterprises to consume generative AI, it is still a challenge to build organizational expertise to the degree needed.
Similar to limitations with finding data scientists, enterprises are now facing a similar challenge with generative AI. This is driving a shift where many enterprises are limiting their use of foundational tools in favor of generative AI functionality built into already-deployed enterprise software. The beauty of this approach is that the vendors can amortize the development expertise and at the same time install guardrails in ways that would be challenging for enterprises.
CIOs focus on the ‘big three’
Just about every software, hardware and services vendor is looking for ways to leverage generative AI into their solution. Due to the sheer amount of computing and storage resources required and the costs associated with running these models, cloud computing is proving to be the lynchpin to enabling generative AI’s explosive growth.
With each rotation of topics (generative AI and beyond), CIOs are centering on how each of these impact the ‘big three’ areas of focus for business. This is further support of the CIO’s shift from being technology-centric to business-centric. In the end, any technology topic needs to tie back to one or more key focus areas for enterprises. The big three are customer experience, employee experience and business operations and supply chain.
Enterprise investments in AI are generally focused around these three key areas. Even large, complicated enterprise solutions like ERP are quickly evolving to leverage this new technology. In a way, it is creating a resurgence in ERP interest. There is a catch. The catch is that a) most of the new functionality is only available in the cloud versus on-premises, b) enterprises face significant risks and costs associated with switching from traditional on-premises versions of ERP to cloud and c) the switch requires a different mindset across the organization. Vendors are doing their part to help smooth this transition and de-risk as much as possible, but it is still complicated.
As if that wasn’t complicated enough, legislators are incredibly busy producing new legislation and regulations around AI and generative AI. One account suggested there are already 126 pieces of legislation out today and more is on the way.
All of this is leading to general fatigue around generative AI. Unfortunately, this is a marathon comprised of a series of sprints and CIOs will need to find ways to address fatigue.
Other topics of interest
There are other topics being discussed among CIOs. Those include a serious discussion about cloud costs and rationalization of cloud resources. In addition, enterprises are still struggling with how to effectively manage hybrid work and the most recent return to office edicts. These are two areas that don’t have a clear direction or answer but are critical to the enterprise’s success.
领英推荐
New content
Lots of new content on both the blog and CIO In The Know podcast. A couple of interesting ones to note are a) the blog post on July 20 showing how accurate Generative AI is for enterprises and b) the interview with CIO David Costar that looks at CIOs from a non-CIO perspective.
Blog
Podcast
What’s your take?
This is just a snapshot and there is a lot to unpack here. Looks for more unpacking in future missives.
Now it’s your turn. I’d love to hear from you and your perspective.
Be sure to subscribe to the CIO In The Know newsletter , CIO In The Know podcast and CxO In The Know podcast to ensure you do not miss an update. Also, if there are others you think would this newsletter valuable, please forward them a copy and encourage them to subscribe.
?
#CIO #CIOitk #CxOitk #newsletter #risk #cybersecurity #CISO #ERP #EnSW #finops #gpt #ai #generativeai #LLM #LLMs #innovation #macroeconomic #podcast #newsletter
Sr. Solutions Architect / Security Solutions Practice
1 年Thanks Tim -- as always, great stuff! Definitely validates all the chatter I've been hearing (for months) in my circles that "AI has solidly dropped into the 'valley of disillusionment' in the hype cycle. But learned lots of new things (the "top 3" for example). Thanks so much for sharing your valuable insights. Appreciate your sage sentiments.
C-Level Advisor | AI & Automation Leader | Solution Design | Systems & Design Thinking | Digital Diagnostician |
1 年Lord knows I'm exhausted trying to keep up with the pace of #generativeAI. Its changing at a pace unrealistic for enterprises to keep up. Good write up Tim.
Do-er of the Difficult, Wizard of Why Not, and Certified IT Curmudgeon
1 年I believe it, inside and out. I'm not in a true, international business at the moment, but one of my biggest worries is whether/how often a GDPR-style right to be forgotten implies curating terabytes (petabytes or more?) of training data and re-training a model from scratch because there's no way to tease out the dirty drops in the ocean of weights and activating functions of the NN itself, It seems like a lot of AI fatigue and corner cases where it breaks down (not to mention hallucinates) are probing the edges of what can be accomplished with unsupervised learning at scale. Humans don't do anything like that, so even if our wetware acts a bit like their software or vice versa, it is trained in a completely different way than we are...
Leader, Builder, Mentor. I help companies scale people and tech to deliver exceptional results to clients.
1 年Tim Crawford, nice write up. Thank you. And so true that the enterprises should be focusing on the "true big 3". Not AWS, Azure, and GCP but "The big three are customer experience, employee experience and business operations and supply chain."
Director, Content Strategy, Gartner Peer Community
1 年Super summary - thanks, Tim!