The Over-Inflation Problem of Gen AI
Last week I wrote a post blaming AI over-hypers for a rising sense of disillusionment. But that's only half the puzzle. Hype-mongers may have raised expectations, but organizations have prevented themselves from getting the benefits they hoped for.
Disillusionment comes from reality not meeting expectations.
Organizations have been expecting dramatic results while ensuring there's no drama in the process. That's like trying to put on a spectacular fireworks display while immediately snuffing out every spark you see with a fire extinguisher.
Everything is done with the best of intentions, but the corporate antibodies ensure that AI doesn't deliver as much as it should.
So, let's explore some of the approaches that—while seemingly prudent—might be limiting the dramatic impact AI can have. And I'll even give you suggestions for tackling the situation better.
?? Unrealistic Expectations ??
Some folks expect AI to deliver jaw-dropping results overnight. That's like expecting a gourmet meal from a microwave - not impossible, but highly unlikely.
?? Set realistic expectations and understand that AI needs time and fine-tuning to really shine.
?? Resistance to Change ??
Like a granny on a rollercoaster, most organizations are uncomfortable with unexpected twists and turns. They're geared up for minor tweaks rather than revolutionary overhauls.
?? Consider how AI might fundamentally improve your processes rather than just adding it to existing workflows.
?? Lack of C-Suite Knowledge ??
The execs might be experts in their fields, but when it comes to AI, it's a new frontier. It's too novel to provide the case studies and best practice documents they typically rely on.
?? Invest in AI education for leadership to help them make informed decisions that balance risk mitigation with opportunity creation.
?? Insufficient Training Investment ??
Failing to train your teams on AI is like giving them a new sports car without teaching them how to drive. Without proper training, employees will stay in first gear and wonder why they're not getting as far as they expected.
?? Prioritize comprehensive AI training programs.
?? Lacklustre In-House Solutions ??
Some companies build their own AI tools that fall short of the standard their employees expect. I've seen them firsthand. Some of them are so hobbled that they're a waste of time.
?? Ensure your in-house AI solutions are close to the capabilities of publicly accessible models if you want people to use them.
领英推荐
?? Short-Term Focus ??
Organizations often chase immediate gains like a cat with a laser pointer, forgetting that AI is more of a long game. The real value comes from looking beyond the next couple of quarters.
?? Develop a long-term AI strategy aligned with your business goals.
?? Inadequate Change Management ??
Implementing AI is a significant change management task. Many organizations flounder because they don't manage the change process effectively.
?? Develop a comprehensive change management plan to guide your AI transformation.
?? Over-Focus on Cost-Cutting ??
Companies often see AI as primarily a way to cut costs. They just want it to do things faster and cheaper. But it's so much more capable than that.
?? The real power of AI lies in augmenting human skills and creating new opportunities, not just trimming the budget.
?? Lack of Clear AI Strategy ??
"Hey, let's see if ChatGPT can help me get through my emails faster" is not a strategy. Neither is "Let's try it out and see what it can do."
?? Start with your business strategy and work out how AI can help you build on it.
?? Thinking It's About Tech, Not People ??
Yes, the technology involves gigatons of silicon nerdiness, but the human element is the most important part of your AI implementation. Humans brief the AI, judge its output, check for accuracy, and refine the results. They're what can take the results from adequate to excellent.
?? Consult your teams and use AI to make their job better rather than soulless
?? Expecting it to Fix Your Organizational Dysfunction ??
AI is impressive, but it's not a magical fixer-upper for deeper organizational issues. In fact, from experience, the process of implementing Generative AI into the workflow often reveals the organizational problems that have been festering for years.
?? Use AI implementation as an opportunity to identify and address fundamental problems.
These are my Top 11 (I pruned the list to get it down to this!) What would you remove to get it down to 10? Or add to make it a round dozen??
LinkedIn [in]structor | Data Science Consulting
1 个月I'd add that a lot of organizations try to skip over the fundamentals like organizing data or diagraming business processes, and instead try to jump ahead to using AI. Trying to implement AI without a business justification or purpose leads to technical debt in the future because it can create additional problems and issues that an organization will need to fix down the road. After all, if they can't measure it, they can't manage it either!
LMS Specialist at Barber National Institute
1 个月I loved this post it was very well done I wonder if you used AI to assist you in creating it? But I agree with every part of the post. Joe
Insightful breakdown! It's crucial for organizations to understand the full potential of Generative AI. Thanks for shedding light on this! ??? Dave Birss
Digital Anthropologist | CMO | I'm in WIRED, Forbes, National Geographic
1 个月What a lot of people don't understand is that CIO's & CTO's understand the hype and that they're real priority is technology debt, not adding new things. Some saw AI as a band-aid to tech debt, but it didn't work and they tore it off. Tech debt is a huge issue for large & small organisations. Most AI projects remain R&D efforts, not actual implementations.
CTO, Gen AI, IT Transformation, Tech Strategy
1 个月Thanks for posting this Dave. Most of your points resonate with what I have discovered during my own Gen AI journey over the last 12 months. I might add the lack of awareness of the potentially changing competitive landscape. This may not apply to all businesses but if your business relies heavily on data, knowledge and IP sourced from third parties then you are in a completely different and new competitive landscape. As a leader it is worth asking, what stops a competitor or a startup from sourcing the same IP, knowledge and data but delivering something more or better using AI and Gen AI in particular. I think leaders need to be constantly assessing on this dimension both internally (in terms of what more they can do) and externally (in terms of who can disrupt them)