Is CHATgpt a friend or an enemy of CI?
I guess if you read any of my columns you know what I am going to say. If you accept your position as an information bot (mislabeled as “competitive intelligence”), CHATgpt may signal the end of your employment. This will be a repeat of what happened when CI platforms started making inroads. A few researchers-cum-“CI-managers” lost their jobs when companies decided to connect their product marketing managers directly with a platform and skip the middle person of “CI” altogether. After all, human information bots (researchers, archivists, etc) don’t add much to what platforms like Klue can do, except cost. (Note: this is by no means an endorsement of Klue. I am using Klue as a clue to the unintended consequence of subscribing to a platform as a source of “intelligence.”)
However, CHATgpt can be a gift, and I don’t look a gift horse in his mouth. Actually, I don’t look in any horse’s mouth. Unless you are a horse trader, why would you look a horse in the mouth? ?OK, I should stop beating this dead horse (what a horrible expression. I am surprised PETA hasn’t marched on the White House to change it for its implied animal cruelty.)
Here is why CHATgpt is a gift
When Jan Herring and I started the Academy in 1999, we had a continuous dialogue about how to define competitive intelligence. Jan, coming from a senior government role, defined it as actionable information. I, coming from a business school, asked: What if management does not act, is it still intelligence? At the time we didn’t reach a conclusion. My question was more along the lines of insolvable riddles such as “if a tree falls in the forest and no one is there” etc. You know, the wiseass academic question without any practical uses.
CHATgpt, at last, provides the definitive answer to this question not because it is a good product but because it is an idiot GIGO product. Ask CHAT to run an analysis on anything and expect garbage of the highest quality. @Glen Lyons demonstrated it by asking CHAT for a solution to electricity production in the US. The actions suggested by CHATgpt were well-regurgitated solar/wind/farts solutions the brain-washed programmers put into its database and omitted solutions such as nuclear, clean coal, carbon capture, etc. In short, CHATgpt is as good as any biased, "progressive" consensus.
So why is this relevant to the definition of CI? Because it demonstrates beautifully and decidedly that CI is not actionable information per se, unless you consider any trite action recommendation to be intelligence. Competitive intelligence is competitive insight- a whole different game, a tiny fraction of the space of all possible actions that has a chance of giving the company a leg on competitors. Or at least a toe.
领英推荐
Insight is never trivial. It is never a formula, and that's why Crony Consultants don’t have insights, they have skillful selling skills only to top management who lost their way. Similarly, CHATgpt can’t by definition produce insights because you can’t program insight into a database. Insight is always conditional on the situation, past data have little relevance, and CHATgpt, therefore, is not an enemy of CI analysts who are the true value-add part of the function.
The remaining “busy work” of compiling mere information can be done by CHAgpt or an insomniac high school student. Same difference.
I understand investors’ enthusiasm. OpenAI is now valued at $30 billion. I think the only delusion is that what happened to EV and Crypto companies is "totally different." The herd produces milk, but no honey. Good luck to those who see it as the future of humanity.
?
Alternative perspective: In a recent Gartner report, the data company recommended product marketing leaders prepare for the future by upskilling their teams in competitive intelligence and recommended ACI for training and certification. As hard as it is for me to agree with anything these giants recommend, this one is not half bad. And it sure beats CHATgpt for competing in an open market (at least the part that is still open as governments worldwide play a bigger and bigger role in choosing winners and losers). I am now waiting for McKinsey and Deloitte to admit in a report that they need ACI to explain to them what CI is about. ?
This essay was written by CHATb-g, not CHATgpt.
If you want to upskill, listen to Gartner and join us in March for the CIP-I upskilling program. Registration - Enroll Now - Academy of Competitive Intelligence (academyci.com)
High-powered innovations in competitive strategy: ForesightSims? simulations, business war games, workshops on strategic thinking, teacher, prolific author including 12 HBR digital articles, nonprofit board member.
2 年As I ponder AI, my brain starts to hurt. Then I ponder, many levels below AI experts, what makes AI (of the ChatGPT variety) NOT work? It occurs to me that we humans take a great deal for granted, subconsciously, about context. I know one person is in Australia, another in Florida, and I'm in Oregon, and those tiny facts have tremendous context. Humans store and use context effortlessly. I know not only about those people in Florida and Australia but also that we understand time zones, and we have conversed on multiple subjects, and we address CI from different directions, and so on. I can understand, in my shallow way, how machine learning works. This is a cat. This is a pillar. This is a caterpillar. This is a Caterpillar. That's categorization. But how can AI grasp context? Is it a matter of an ever-larger database of Oregon, Florida, and Australia? Of Earth? And, of course, we humans get context wrong from time to time. As John Allen Paulos put it in one of his books about innumeracy, the odds that a person in Spain speaks Spanish are very different from the odds that a person speaking Spanish is in Spain.
Systems thinker, life-long learner, facilitator and mentor. I love the challenge of helping people see the big picture to make big decisions. Principal and Founder of Caldgargan & Associates.
2 年How is it that everytime I try to use CHATgpt it puts me on a waiting list? So I used the tool that Bob Gourley (OODA LLC--a great intel analyst I worked with on the Joint Staff in a different life) Unrestricted Intelligence (https://unrestrictedintelligence.com/) that uses the same AI algorithms as CHATgpt, the most important question of all: What is the meaning of Life? And it failed. The answer of course is 42 not "The meaning of life is a complex and often deeply personal question". Lets go back to using our brains and not code.
Systems thinker, life-long learner, facilitator and mentor. I love the challenge of helping people see the big picture to make big decisions. Principal and Founder of Caldgargan & Associates.
2 年Oh My we are thinking alike. Its because you trained me. I did my own sarcastic version of AI last week and in this shameless plug posting it here. But if it were not for Ben drilling into my thick skull to THINK, I would not be the analyst I am today. https://www.dhirubhai.net/feed/update/urn:li:activity:7028880018407624704/
?? Senior Sales Leader | Revenue Growth & Competitive Intelligence Expert | Driving Success in SLED, Manufacturing, B2B & Education ?? Territory Expansion | Strategic Sales | High-Impact Client Solutions
2 年Spot on! Reading my mind.
?? ???????? ???????? ??????? 20 ?????????? ???? ???????????????????? ???? ?????? ????????????????????, ???????????????? ?????? ???????????????? ???? ???????????????? ?????? ????????????????????
2 年Luigi Simone Menichino