Awaiting AI's big bang: When will technology stun the world?

Hi there ??,

Welcome to TomTalks??!

Lately, I've found myself daydreaming about when AI will deliver something so groundbreaking, so undeniable, that its full potential becomes clear to everyone.

Will it be next year? Five years? 10? ??

It's hard to imagine it taking that long, but on the other hand, AI is already doing groundbreaking things, like:

• predicting the chemical structures for millions of all-new materials ??
• revolutionizing healthcare diagnostics ??
• generating new types of medicines ??
• copy human skills in real-time without prior training . . . ??

. . . and yet, deep appreciation for AI capabilities is still concentrated in circles of technologists, researchers, and a handful of other professional and/or creative types.

When will that change, and how will it happen?

Will it be a good moment, like an AI winning a Nobel Peace Prize? ??

Or a bad one, like an AI declaring itself sovereign and attempting to inflict harm within its sphere of influence? ??

?? Q* and Gemini ??

Spurring my thoughts on this topic has been OpenAI's reported development of a new AI system called Q* (pronounced Q-star), and Google's newly released Gemini upgrade to Bard.

It's worth noting that there's disagreement about whether Q* is anything special, and Gemini was just cracked open yesterday, but the hype around both systems is that they feature (or will) advanced problem-solving capabilities that transcend what we've seen with GPT-4, much less any other chatbots.

Specifically, there's a possibility that these systems can solve mathematical equations, plan tasks before executing, and perform analytical research. These might not sound like big leaps forward unless you already know that these are areas where AI has struggled quite a bit.

?? Practically speaking ??

Here's a way to think about it that stays above the technical jargon:

However good Gemini is, and whatever Q* turns out to be, even better systems are on the way. And the better they get, the less hand-holding humans will need to do; the more independent AI will be able to operate.

Instead of simple jobs, we'll probably ask it to do complex ones:

Strategize. Plan. Coordinate. Integrate. Execute.

?? New user experiences ??

In all likelihood, improved capabilities will lead to vastly different ways of interacting with AI, too.

Instead of linear conversations with chatbots, we'll have the option of using multi-modal, interactive environments where we'll be able to set broad objectives and monitor as AI actively sources, manages, and presents diverse types of data, even forms of rich media, in customizable dashboards or via integrated tools and databases.

Today, for example, AI has a lot to offer to, say, a digital marketer.

But in years (or months?) to come, it will offer a genuine superpower, having the ability to strategize a campaign, create content, manage postings across platforms, and analyze feedback—all visible and manageable through an interactive interface.

I still don't see it replacing digital marketers . . .

. . . but it's changing (and will continue to change, even more drastically) what the profession entails.

It might seem pointless to try and visualize the experience around something that doesn't exist yet, but the opposite is actually true.

Not knowing what the future will look like, physically and experientially—and how that will change our ways of working on and thinking about problems—is one of the reasons it's so hard to prepare for that future. Creating a mental picture can also help us anticipate some of the challenges we'll probably face, and preemptively design with them in mind.

?? New and evolved applications ??

I told ChatGPT what I was thinking about and asked it to imagine some of the things that systems like Q* and Gemini might be exceptionally good at. Here's what it said:

• Complex Project Management: These AI systems could excel in managing large-scale projects, involving numerous variables and stakeholders. They could optimize resource allocation, track progress, predict and mitigate risks, and ensure that project milestones are met efficiently.
• Financial Analysis and Forecasting: They could be adept at analyzing financial trends, conducting risk assessments, and offering investment strategies by processing vast amounts of financial data and market trends.
• Healthcare and Medical Diagnosis: Advanced AI tools could assist in diagnosing diseases by analyzing medical data, patient history, and current research, potentially identifying patterns that are not immediately obvious to human practitioners.
• Environmental and Climate Modeling: They could be crucial in modeling climate change scenarios, analyzing environmental data, and proposing sustainable solutions to complex ecological problems.
• Legal and Compliance Analysis: These tools could aid in legal research, case analysis, and ensuring compliance with ever-changing regulations in various industries.
• Educational and Training Programs: They could create personalized learning paths for students, adapting to their learning styles and pace, and providing comprehensive educational content.

I'm sure this is only a cube off the proverbial iceberg. ?? Just like with the internet, if your organization isn't implementing or at least working on a dedicated strategy for how to identify and integrate game-changing AI tools both now and in the future, it's already fallen behind.

AI and its influence are growing exponentially, creating a world filled with deep opportunities and even deeper unknowns. TomTalks ?? is a weekly exploration of the benefits, risks, and costs of AI adoption, featuring brief but crucial conversations with AI experts and global business leaders.

Want to be a guest on a video podcast for TomTalks? Please DM!