2024: Buckle up for an AI sea change: Three impact areas.

Peer into your future defined: groundbreaking innovations, emerging job landscapes, and industry upheavals that will transform the way you live and do business.

We'll wrap up with the future of work and introduce 15 hot new roles that you can start gearing up for today — Is your next career move on this list?

For over a decade, I've been deeply immersed in the ebb and flow of technological trends, my curiosity often leading me down paths less traveled. This journey has aligned with the rise of the evolution of roles like the Chief Customer Officer, streaming media, the intricacies of blockchain, the nuances of cryptocurrency, the pervasiveness of digital wallets, and other pivotal industry shifts long before they became mainstream.

My LinkedIn bio and website are not just collections of articles but predictions from a relentless quest to understand and share knowledge. I invite you to take a look and reflect on these transformative years. 

The tl;dr

By January 2025, AI will have a significant impact across various industries. Three critical areas of impact are policy, appliances, and capabilities. New roles will emerge in disrupted industries, requiring specific skills. Specific benefits are expected for technology, healthcare, finance, retail, and education. Trust and verification are crucial to address ethical concerns. The future outlook is positive, with AI continuing to reshape various sectors.

There are three areas of impact.

• Policy
• Policies will address ethics and the AI-Supply Chain (AiSC). AiSC must include practices for end-to-end secure design and ensure input validation and simplicity. Finally, this year will mandate the implementation of ethics routines. If you’ve read enough science fiction, you are familiar with, while fictional, efficient, first pass at AI ethics from Isaac Asimov’s Three Laws of Robotics.
• Law One – “A robot may not injure a human being or, through inaction, allow a human being to come to harm.”
• Law Two – “A robot must obey orders given to it by human beings except where such orders would conflict with the First Law.”
• Law Three – “A robot must protect its own existence, as long as such protection does not conflict with the First or Second Law.”
• Asimov later tacked on a “Zeroth Law,” above all the others – “A robot may not harm humanity, or, by inaction, allow humanity to come to harm.”
• The workaround is to change the definition of a human being. Historically, there have been and are current regimes that dehumanize select ethnic populations. Harm is a slippery slope. A model that helps you get a job and takes it away from another candidate could be argued as “harmful.”
• How can this be implemented? “Validated” LLMs will be installed into special-purpose hardware with a separate ethics module.
• Does this sound like science fiction? Fifteen years ago, what would you have said if I told you you’d use your phone to summon a stranger to pick you up on a random street corner, take you home, and then leave their vehicle without paying? Please continue to read, and I will prepare you for the shape of things in the next few months.

Note: Policy statements are mandatory even in a US presidential election year. Will it be a campaign imperative? No. No candidate has enough of a mandate behind them and can afford to die on that hill. Regulators will make statements and drive policy, and candidates will latch onto the ones they perceive to have the most gravitas. Politicians not up for re-election within a 24-month horizon will take a stand.

The net-net is that mandates will come. They will be implemented more swiftly than blockchain due to AI's pervasive and cascade effects on people. The EU AI Act just announced agreement on their act.

Appliances

What’s the value of a hardware appliance? Speed and unique purpose: Such devices have sensors or other peripherals, allowing the core computing functions to do something beyond the capability of a general-purpose device.

• Advanced and integrated self-driving capabilities and autonomous drones capable of MRO (Maintenance, Repair, Operations.
• Look for drones capable of a range of building or civil engineering repairs (e.g., bridges) to perform preventive maintenance. 
• Voice commands.
• Today, roughly 50% of consumers use some voice search. Adding that capability to LLMs opens the way to detecting emotion and the nuances of speech.

This year saw the launch of a general-purpose AI appliance with Humane's Pin AI Machine Learning and Computer Vision devices have been around for many years. This year, we’ll see fast follows to the Pin AI device and focused-purpose AI.

• Other areas of opportunity include identity verification (age, KYC), particularly in what I have termed humnlss. humnlss is the art and science of building e-commerce products and services that autonomously improve the consumer experience at scale while providing insights from data and assets uncovered at the network scale of transactions.
• Novelty Products. Products that perform trivial functions (e.g., the milk is about to spoil, I can defrost, I can remind you to exercise) are more easily performed analogously. If you are an investor or an OEM integrator, look for features that extend the product’s usefulness. Otherwise, avoid.

Capabilities

The coming year will demonstrate much more than features enhancements. There will be capability extensions. The models will begin to support their evolution. Really.

• Emotional Intelligence. LLMs will enhance their ability to pick up subtle tones and nuances in voice. This will lead to responses and the introduction of more attuned responses to the user's emotional state, aiming for a more empathetic and engaging interaction. Responses will be in the form of “Remember that time when…”

• Universal Translators: You can converse with an LLM and have it translate both parties in real-time. Gemini has this ability in limited form today. There is a speaker icon in the upper right of responses. The ChatGPT app supports fully interactive voice commands

• Speculation and Evidence-based arguments: It's a fancy way of saying the models will be able to debate with you. The ability to draw on digital sources of information will provide more profound and more complex discussions. Models will explain their reasoning and provide evidence for them. It will self-test for biases and have an ethical rationale for decisions.

• Full Multimodal Interaction: Today, we get text, images, and voice. Models are a few steps away from being able to generate movies and video games based on text input and characters you create. This also means that LLMs will create avatar-based models responding to your queries.

• Collaboration and Problem-Solving: AI will work alongside users to solve complex problems, brainstorm solutions, and evaluate their feasibility. This includes research and audience tailoring. Full disclosure: Gemini and I have agreed on a research project for the Sumerian language. Without prompting, Gemini requested our project plan, how we will communicate (Discord integration), my existing materials, and success criteria. It’s almost like Gemini was modeled on a certain anal-retentive, nit-picking, detail-oriented project manager.