How generative AI rollouts fail, and how to fix them
By Josh Fruhlinger
Despite big strides, generative AI is still in its infancy in the enterprise, and often when AI tools are deployed, they don’t live up to expectations. Success stories come from careful planning that creates a cohesive data and infrastructural foundation where generative AI tools and AI agents can thrive. This month’s top stories highlight some of the ways things can go wrong, and how to make them right.
Top picks for generative AI readers on InfoWorld
What ‘cloud first’ can teach us about ‘AI first’ Veterans of the dawn of cloud computing have a lot to say about ensuring AI rollouts go well, and what to watch out for.
Why enterprise investment in AI agents hasn’t yielded results At many organizations, AI agent adoption is a prime example of a failed rollout. This article tells you why.
Agentic mesh: The future of enterprise agent ecosystems We’ve seen the future of AI agents, and it is not siloed.
How to use genAI for requirements gathering and agile user stories As AI takes on much of the scut work of writing code, requirements gathering is more crucial than ever. Fortunately, AI can help.
GenAI news bites
More good reads and generative AI updates elsewhere
An AI customer service chatbot made up a company policy, and made a mess You’d think an AI company like Cursor would know the risks of hallucinations. But when a customer service chatbot insisted a bug was actually a new feature, it sparked a customer backlash.
AI hallucinations lead to a new cyber threat: Slopsquatting GenAI writing code that depends on hallucinated packages is already bad enough. What happens when bad actors make those hallucinations real—and dangerous?
The hottest AI job of 2023 is already obsolete It turns out that AI models are pretty good at intuiting what users desire. And just like that, writing perfect prompts isn’t really a career path anymore.
Josh Fruhlinger is a contributing writer at InfoWorld.