Editor’s note: Tom Snyder, executive director of rapidly growing Raleigh-based RIoT and a thought leader in the emerging Internet of Things, is the newest columnist to join WRAL TechWire’s list of top drawer contributors. “Datafication Nation” columns are part of WRAL TechWire’s Startup Monday package.
+++
RALEIGH- There is tremendous hype and hysteria surrounding Generative AI, and the risk it poses to many job categories and to complete industries. But should we believe the hype? Are all those jobs going to disappear tomorrow? The short answer is “no. Anyone who is familiar with the Gartner Hype Cycle knows that every hot technology has to survive the “Trough of Disillusionment” before eventually rising to long-term prominence. But there is one area where I think an atypically rapid technology “death” may occur.
As I have described previously in this column, the Data Economy, underpinned by IoT, is all about moving from legacy information-based tools to real-time, data-driven automation. The Internet Age was powered by Search.
Google launched in 1998, along with a few competitors (remember Lycos? Altavista? Ask Jeeves?). And while those competitors were not as successful as Google, without question, Search was the killer app for the internet, driving everyone to use it and eventually make it a daily part of their lives and businesses. It was, and remains today, the most important application of the internet. But that is likely to change quickly.
Search puts all of the world’s information at our fingertips. It is a tool that we used to augment our own knowledge. We “Google” everything, whether it is searching for hotels in Barcelona or movie times at a local theater or historical dates for a term paper. Search is so significantly impactful that super-search platforms emerged in niche areas. Kayak, for example, runs numerous parallel searches and consolidates results. But Search was always limited to finding information efficiently.
‘Datafication Nation:’ Exec director of RIoT to explore data’s growing impact on our lives
Generative AI takes us a major step further. Now, with a well-written prompt, we can both SEARCH and DO. (NC State was close with their “Think and Do” motto). No longer do we have to simply search for hotels in Barcelona. Generative AI allows us to instruct, as if to a digital assistant, “find and book a hotel that meets the criteria I have outlined”. “Reserve a ticket at the nearest theater for this evening and book a rideshare to get me there”. The tech begins to Automate the steps we used to conduct after we did a search.
DEFINING GENERATIVE AI
“Generative artificial intelligence or generative AI is a type of artificial intelligence system capable of generating text, images, or other media in response to prompts. Generative AI models learn the patterns and structure of their input training data, and then generate new data that has similar characteristics.” – Wikipedia
Of course, the examples I give above are not yet available (that I’m aware of). But they will be very soon. Generative AI as a foundational platform is already mature enough to do the work after the search. One of the most common applications already is to use Generative AI tools to create instructions for you. “Create a dinner plan based on the ingredients in my refrigerator” is an oft-cited example. Generative AI knows how to automate today.
Companies will soon release “purchasing” integrations, “scheduling” integrations, “approval” integrations and all manner of others to connect Generative AI tools to our bank accounts, calendars, personal data, preferences and allowances. With these integrations, the power of the AI to truly be an assistant is unlocked.
It will seem quaint that we used to search for information, and then have to do the work associated with clicking the responded links, reading the information, making decisions and taking actions that could instead have been all automated by an assistant that knows our preferences and desires. While it will take years before we have the robotics and automation in every home to actually cook the dinner from the example above (Jetsons anyone?), the work we already do today on digital platforms like booking travel or purchasing goods for delivery have few barriers to immediate adoption.
Report: Artificial intelligence market growing at 37% a year
I am hesitant to go so far as to declare Search is Dead – the initial title for this piece – for one reason. Not all data sets are available to the Generative AI tools. Increasingly, data owners are going to recognize the value of their data and set up paywalls or other business models around data usage. Reddit’s recent decision to monetize their API is an early example.
This capitalist exploration for how to monetize the tech will create data gaps that limit capabilities for some use cases of prompt-based automation. There will be fragmentation in the industry as new data access, data licensing and data partnerships emerge. We’ll additionally see a period of free search and paid prompt-based automation.
Ultimately, automation is the “killer app” of the Data Economy, and the money will work itself out. The next Google will create ways to monetize prompt-based automation in the manner that drives widespread adoption like search created 25 years ago. But it will happen faster.
The End of Search is on the horizon.
‘Datafication Nation:’ Internet of Things as a descriptor is changing – but not its importance
The post Generative AI means the looming end of search as we know it – here’s why first appeared on WRAL TechWire.
Generative AI means the looming end of search as we know it – here’s why first appeared on Web and IT News.