Back to Main
This Week in AI — Taming AI Hype by Having a Plan
AI

This Week in AI — Taming AI Hype by Having a Plan

October 17, 2023 | By: Dave Isaac

Headlines You Should Know

 

Beware the Hype Train. Charting Your Own AI Course is the Way to Go.

A steady stream of rosy press about new AI tools and capabilities has flooded the news cycle in recent months. Even top-tier tech reporters have written glowingly about the AI revolution after getting a trial run of products before they’re released to the public. The undertone to it all is that if you’re not getting good results like the demos often show, you may be the problem. 

Fear not. Your intuition is correct. Bard really does stink — even Google insiders think so. These AI tools don’t come with instruction manuals, so to an extent, you have to be your own teacher. And the models themselves are changing, so what you learned last week may not be true next week. 

All this is to say that the AI hype train moves recklessly fast, but your AI story will be different than anyone else’s. Start experimenting with AI tools on a couple of tasks you could use help with or that take you too long. Use that time as an investment to find out how you can make your work life easier. For more direction from a particularly good voice of reason, and validation that the AI world is uncertain and moving fast, this FAQ from Wharton professor Ethan Mollick is a great resource. 

Do Companies Have an AI Adoption Plan?

AI hype isn’t just infecting individuals — companies are ready to start their AI excursions whether they have a roadmap or not. Gartner says 90% of companies will have adopted AI by 2025, yet only 9% “have an AI vision statement with established principles in place. More than one-third of companies had no plans to create ground rules.” 

Just like you need to be thoughtful about the words you choose in a prompt, you (or your company) should have a strategy behind AI use and clear boundaries for what processes or data should be off-limits.

Elsewhere …

Tips and Tricks

🎨 DALL·E 3 Arrives in GPT-4

What’s happening: Last week we covered ChatGPT’s ability to take images as inputs, and now it can produce them as outputs, too. DALL·E 3, OpenAI’s image generator, had been available through Bing Image Creator and is now available within ChatGPT for Plus and Enterprise users.

Why it matters: The new integration means it’s easier to access DALL·E 3 if you’re already using ChatGPT, and you have a few more options. For instance, Bing Image Creator would only create images that were 1024×1024 pixels, but with ChatGPT you can specify a different aspect ratio. You can also make changes to iterate off one image instead of starting from scratch each time.

Try it out: Click on “New Chat” and hover over the “GPT-4” box to see the different capabilities. You’ll now see “DALL·E 3” at the bottom of the list, beneath “Plugins.”

Quote of the Week

“Generative AI is more than a fleeting trend or mere hype. It is a transformative technology with far-reaching implications and business impact. With ethical and responsible implementation, GenAI is poised to reshape industries, changing the way we work, play, and interact with the world.”

— Ritu Jyoti, Group Vice President at IDC, which forecasts GenAI spending to reach $143 Billion in 2027