We conducted a survey about a week ago with a hundred CIOs and CTOs who said that AI is going to be crucial during the next two years. But, they also said in the same survey that they don’t think their teams will be able to drive that change. What’s your perspective on how IT Executives can better prepare for it?
VP of Product Management, 10,001+ employees
There is a need for cross skilling existing teams and executives with the new age technologies and possibilities Offered by the same. There are programs by the universities as well as online courses that could be undertaken by the executive teams. What works best is KPI driven approach for cross training of executives and team. Often bringing in an external expertise including consulting teams can also help.
Founder/CTO in Hardware, 11 - 50 employees
Finding the talent to do the work is going to be a big issues. Data Scientist and big data resources are getting top dollar these days. The other area is companies have to be able to understand what is a good project for AI and what is just a straight Data Analytics problem. I have seen companies say I need AI when the reality is the do not. Lastly, do they have enough data to do the training of the models. Advisor | Investor | Former CIO in Services (non-Government), Self-employed
Pat -- you nailed the question. It's not just the question of talent, but the training datasets. Without the appropriate training datasets to train the models, it's like having malt and yeast without the recipe for brewing good beer.
VP IT (CIO role) in Healthcare and Biotech, 1,001 - 5,000 employees
Hiring qualified and affordable resources is a challenge, especially here in the bay area. Some consulting companies marketing AI are just using predictive analytics/algorithm that already existed 15 years ago but that's not true AI to me. Finding implementations of successful use cases of true AI is still a challenge although I expect this will improve rapidly with increased adoption.CTO in Software, 2 - 10 employees
Two things:
1) Realize that AI and ML are a symptom of broader trend, *becoming data driven*. Models are a part of that, but not the only part. Successful teams will invest in training both on the technical side (AI), and on the culture side. The quest to make data-driven decisions starts with the culture. Does your company make decisions with the aid of models, visualizations, notebooks, and hard data? Or do decisions fall from the sky on stone tablets? Marissa Mayer famously said that she discourages people from saying "I like X" in meetings--and rather encourages them to bring data that support or refute specific points of view.
2) Invest in data infrastructure that allows *the entire team* to participate in the process of interpreting and making decisions based on shared data. Too often data are siloed across systems and there is no single location where analysts, developers, and business leaders can get on the same page, reason about the world, and document their decisions.
CIO in Education, 1,001 - 5,000 employees
We need to start training some people internally, and leveraging tools that we may be using elsewhere (e.g. Microsoft Q&A maker) to begin some initial steps with existing resources. Director - Transformation in Software, 1,001 - 5,000 employees
AI is the end result of well implemented BI and ML programs. however, the technology change from BI to ML to AI are step changes from one another and require different levels of investment in talent, skill set, technology, infrastructure and operations. AI and ML will also drive changes to core business operations at the company based on the application of AI. Legacy IT teams will most probably be hesitant to implement AI as most leadership teams treat AI as a buzz word and view it as the \"next shiny toy\" to show off rather than view it as a serious restructuring tool. Implementation of AI requires transformation thinking and necessitate some business model changes. Once leaders realize how difficult, expensive and time consuming it is to put an enterprise grade AI program in place, some structure tends to fall in place and use cases will be rolled out slowly. IT executives can prepare for the change by taking a long, hard look at their company's business model and identifying key focus areas where AI may delivery reasonable results with existing resources. They may then expand their data programs to augment implementation of a sustainable AI program.Chief Information Technology Officer in Finance (non-banking), 1,001 - 5,000 employees
Find a real need rather than inflating the hype of adding AI unnecessarily into the environment, as it just simply creates a bubble of unstable frameworks that will not lead to a proper development of its true benefits.VP of Global IT and Cybersecurity in Manufacturing, 501 - 1,000 employees
AI and RPA AUTO can help the business drive positive business outcomes with various business processes and other tasks, both strategic and tactical, but first the business have a clearly mapped out problem to solve for. Otherwise it may result in developing solutions for problems/situations the business may not have just yet. COO in Healthcare and Biotech, 5,001 - 10,000 employees
CIO’s need to begin to prototype and pilot AI and ML projects. When possible, they need to look at partnerships. When it comes to internal skills and team building, they need to anchor around a dedicated leader who has experience.
Content you might like
CEO in Services (non-Government), Self-employed
Using AI tools 2-3 a week. Use cases: -summaries of content
-slide outlines
-abstracts
-citations.
-Beauti.Ai for slide preparation
-Chat GPT 4
-Styluschat
Privacy11%
Employee Dissatisfaction31%
Creating a digital wall with customers18%
Security/Compliance18%
Validity/reliability of responses19%
Other (please note in the comments below)0%
No problem, bring it on3%
133 PARTICIPANTS
Director of IT, Self-employed
ChatGPT has helped many of my users, including myself, write more effective emails. JavaScript and Node JS35%
Python61%
PHP3%
158 PARTICIPANTS
Director of IT in Education, 10,001+ employees
Learning, Pseudocode, Code completion, quick answers
In terms of building teams, as I said earlier, it’s going to be a challenge. Hiring data scientists or machine learning engineers is hard enough. So, my advice is to look at companies who have done this, won customers, have case studies, have a vision about building out these capabilities as entire platforms and are focusing their resources and intellect in doing this.