At first, it seemed as if it came out of nowhere, like an asteroid hitting the center of the cyber-ocean. You know you’re going to feel it. And after reading this review of my experiment, you’ll also feel why this time, the rippling effects of this tech tsunami leave nowhere to hide.
It’s called ChatGPT-3.5. It’s the latest natural language Generative Artificial Intelligence texting chatbot, a type of AI that creates new content from learned patterns of prior content and training. It’s a new branch of AI, unlike traditional AI, which looks for patterns and trends in data pools to make all-important predictions. Generative AI tries to interpret the question as deeply or broadly as you ask it to and respond in a natural written context.
This human-like generative AI platform was created by San Francisco start-up OpenAI, which recently received a $10 billion investment from Microsoft whose own AI experiment Turing-NLG was falling behind. That’s a big tip-of-the-brim even for tech these days. So. Is something big on the boil?
The technology is not new. Several other tech companies have cooked up similar versions of machine learning AI platforms in the last few years including some heavy hitters you may know:
- Google’s BARD
- Facebook’s RoBERTa
- Amazon’s SageMaker GPT-3
- IBM’s Watson Natural Language Processing
- Wolfram’s Mathematica
- Microsoft’s Turing-NLG
But none got much traction. Perhaps the most notable was a big tv game show star. Do you remember IBM Watson as the middle player on Jeopardy back in 2011? That was early AI doing its thing and we loved it, watching heralded Jeopardy champs get their minds clobbered by a creepy over-calculating black box with swirling blue lights perched in-between redundant humans. Lol. Still. I admit creepy or not it was utterly mind-blowing to watch.
And then there was a new leader. In 2017 Google’s DeepMind Alpha-Go beat the world’s best GO player, followed in 2018 by Alpha-Fold which correctly predicted complex DNA protein fold structures. From there the sky was the limit, and there was no stopping it.
But then like a glorious sunset it all faded. At least from the headlines. Mostly because the use-case was too narrow and it was expensive to engage and access terabytes of data at any instant. That meant access was largely only affordable to super-sized institutions with large piles of data, and deep pockets to mine. Nonetheless, Watson and the others needed a way to monetize the billions invested in AI and they found it. But meanwhile in the cradle the foundational birth of generative artificial intelligence was looking for a breakthrough.
It would take another 4 years before that would happen in late 2022, and to say it’s a breakthrough means it still costs more to operate ChatGPT which has yet to turn a profit.
So why is this big news?
Originally this article was outlined to sample CEOs in 2023 and discuss their reactions to the Fed’s recent interest rate moves to clamp down on inflation. But then ChatGPT went viral and dropped the mike on my toe.
Suddenly there was a new free-access natural large language model text bot on the scene released by OpenAI with enough data stored to make its responses incredibly useful across nearly any subject, and that exploded the internet. It captured global attention instantly and in less than 60 days miraculously accumulated more than 100 million users, a record-smashing achievement 4x faster than the reigning champ, Tiktok!
Flash Headlines! Students cheating on college essays written by ChatGPT.
It’s been all over the news, headline after headline. I’m sure you’ve heard about it.
In less than 2 months ChatGPT has passed the legal Bar exam, medical licensing exam, MBA finance exam (ouch!), a physics exam, written books, essays, songs, poetry, and complex computer code for programmers, cutting a day’s work down to mere minutes. In that time since however, it’s also making some people nervous about their jobs. Teachers, writers, content creators, coders, scientists, professors, poets, artists, you name it, and even YouTube influencers are suddenly uncomfortably stunned by their flickering screens, like humans in highlights. Something has come for us.
For the first time in a long time some new mind-blowing technology has yet again come for our jobs. This time more than the AI automated robots that now make the AI automated self-driving cars we buy from the AI automated websites we search. Heaven, right?
The good news is ChatGPT is at this point “knowledge over movement.” It’s trained on unprecedented amounts of historical data and that’s what makes ‘most’ of its answers so comprehensively accurate. It pulls from millions of lines of texts in books, articles, research papers, forums, social media blogs, websites, Wikipedia, and millions more content sources and makes it all together accessible. It then tries to create entirely new content pulling from this giant horde all at once. And the results can be jaw-dropping, and for me a bit dehumanizing too.
Rather than using AI on a narrow institutional data set to find efficiency patterns and make predictions like Watson-Health AI for example, ChatGPT is simply trying to predict the next word in an answer sequence from any related source within its data realm (thru 2021). Imagine a subject-tasked research assistant enters a library and minutes later returns with a comprehensive report for you. Amazing right? Now, replace that human with an AI bot and teach it to learn from each trip to its own vast digital library and Boom! The newest HAL 9000 (2001 A Space Odyssey) computer is born. And we know how that ended.
But let’s stay curious. How does it produce professionally smooth and comprehensive answers?
It’s all in the question you ask, according to Search Engine Journal expert Roger Montti in What is ChatGPT And How Can You Use It?
“What sets ChatGPT apart from a simple chatbot, says Montti is that it was specifically trained to understand the human intent in a question…”
In other words, the more descriptive the question, the more detailed the answer? That makes sense.
So – What’s next – Is ChatGPT friend or foe?
The potential impact of such a human enhancing productivity tool is not easy to digest when your job is abruptly on the line. Does anybody mourn the tragic loss of fired mimeograph manufacturing minions when Xerox introduced the world’s first paper copy machine back in 1959? No.
Or how about parking-meter coin-collectors when digital parking meters sent them packing under a freeway overpass? According to a 2017 comprehensive McKinsey research study ‘by 2030, around 39 million Americans could lose their jobs due to the shift toward automation.’
So here we are again. The next industrial revolution productivity wave has arrived. And with it yet again a cyber Trojan Horse. And for many who can live with it as a simple productivity tool rather than a job eating cyborg, I suppose ChatGPT is likely to be well embraced. But. If not, this new know-it-all HAL 9000 copycat may have to go.
You knew this was coming. Just relax. I want to help you.
According to Business Insider Magazine the first signs of job disruption will indeed come in the form of an innocent helpful productivity tool. You be the judge. They suggest we not be afraid. ChatGPT is only here to help.
- Use ChatGPT in Search as a Google alternative
- Use it to write essays, speeches, cover letters, songs, scripts, even employee evaluations
- Use it to analyze vast amounts of data
- Use it for scheduling tasks, planning, and time management
- Ask it how to start a business, make strategic decisions, and support customers
- Turn it into a computer coding assistant
- Use it to apply for a new job or to negotiate a raise
Sounds a lot more than just a productivity tool to me. I love the last one on the list. This one made me smile. In other words, even if ChatGPT terminates your job… it can help you find a new one? Does that include me? Surely not my job as a top-rung professional business advisor, I wouldn’t think, right?
Hmmm. What if I ask my new know-it-all friend that question:
Which industries are most likely impacted by ChatGPT? Here reprinted is the answer:
[ChatGPT: The industries that are most impacted by the use of language models such as myself are:
- Customer Service and Support
- Marketing and Advertising
- Finance and Banking
- Transportation and Logistics
- Human Resources]
Ok. That’s a lot. Can you give me some examples?
[ChatGPT: Here are some examples for each of the impacted industries:
- Customer Service and Support: Chatbots powered by language models can be used to automate customer service inquiries, reducing response times and improving customer satisfaction.
- Marketing and Advertising: Language models can be used to generate creative and personalized advertisements, product descriptions, and social media posts.
- E-commerce: Language models can be used to provide product recommendations and answer customer questions about product specifications, availability, and shipping information.
- Healthcare: Language models can be used to assist with medical diagnosis, treatment planning, and patient education.
- Finance and Banking: Language models can be used to automate financial reports, analyze market trends, and provide personalized investment advice.
- Education: Language models can be used to generate personalized study plans, answer student questions, and provide instructional material.
- Transportation and Logistics: Language models can be used to optimize delivery routes, track packages, and provide real-time traffic updates.
- Human Resources: Language models can be used to automate the recruitment process, assist with onboarding, and provide support for employee benefits and payroll inquiries.
Please note that these are just a few examples and the applications of language models can vary greatly depending on the specific use case.]
That’s a good answer, I quipped back but with a confident unwritten huff. Then it hit me as I re-scanned the list, I realized the damage is going to be to a wide swath of people across a wide spectrum of jobs, and ChatGPT couldn’t care less about mine.
So I got right down to it and asked: If I’m a CEO, what should I pay most attention to here?
[ChatGPT: There are several reasons why OpenAI technology is important to CEOs:
Here are some examples:
- Increased Efficiency and Productivity: For example, a manufacturing company can use AI-powered robots to automate the assembly line, increasing the speed and efficiency of production.
- Improved Decision Making: For example, a retail company can use AI to analyze sales data and customer behavior, identifying trends and patterns that can inform decisions about product pricing and inventory management.
- Competitive Advantage: For example, a transportation company that uses AI-powered logistics optimization can reduce delivery times and costs, providing a competitive edge over rivals.
- Cost savings: For example, an energy company can use AI to optimize energy usage in buildings, resulting in significant cost savings.
- Personalization: For example, an e-commerce company can use AI to recommend products to customers based on their browsing and purchase history, leading to increased sales and customer satisfaction.
- Predictive Maintenance: For example, an airline company can use AI-powered predictive maintenance to predict when an aircraft component is likely to fail, enabling the company to schedule maintenance at the most cost-effective time and minimize downtime.
- Identify new opportunities: For example, a financial services company can use AI to analyze market data and identify new investment opportunities.
- Predictive modeling: For example, a healthcare company can use AI to create predictive models to forecast future patient needs and allocate resources accordingly.
- Cybersecurity: For example, a company can use AI to monitor their network for suspicious activity and respond to potential cyber threats in real time
Wow! It’s not Shakespeare, but I do fear thy able apprentice scribes henceforth be put afield… aka in basic English it’s clearly only researchers looking for a new job – Surely us few top-drawer A-list strategy consultants have little to fear as well? Right? HAL?
Are you listening? How about a real test? You think you’re so smart, prove it! Show me the money! Your future vs mine! – Ready, go!
How can ChatGPT specifically help the CEO of a 10-location restaurant in SoCal save money in 2023?
[ChatGPT: Here are a few ways that a 10-location restaurant could potentially save money in 2023:
- Streamline operations: Look for ways to streamline operations and reduce waste, such as using a more efficient ordering and inventory management system, or cutting down on the use of disposable products like paper napkins and plastic utensils.
- Reduce food waste: Implement strategies to reduce food waste, such as portion control, menu planning, and monitoring expiration dates.
- Energy efficiency: Invest in energy-efficient equipment, such as LED lighting and energy-star rated appliances, and take steps to reduce energy consumption, such as turning off equipment when not in use and reducing heating and cooling usage during slow periods.
- Staffing optimization: Evaluate staffing levels to ensure that they are optimized for each location, and consider implementing flexible scheduling to reduce labor costs.
- Marketing and promotions: Use data and analytics to target your marketing and promotions to the right audience and reduce costs. Consider alternative marketing channels, such as social media, email marketing, and influencer marketing.
- Supplier negotiation: Negotiate better prices and terms with suppliers to reduce the cost of goods sold. Consider consolidating orders to negotiate better deals and reduce shipping costs.
- Technology implementation: Implement cost-effective technology solutions, such as cloud-based point of sale systems, online ordering and delivery platforms, and online scheduling systems.]
And the winner is? Yikes! What’s stopping all my customers & clients from asking their own DIY business questions? Where does that leave me and the millions of us minions still on the job?
Good question. And it’s only the beginning. ChatGPT itself thinks it’s only here to help. In fact, according to reports the tidal wave of new generative AI natural language products coming online this year is unprecedented. And as I said, it will find you.
But what if you can’t re-skill that quickly? Is this the END or the beginning of work as we know it?
Maybe it’s the beginning of the end. And what’s more some will go first according to renowned astrophysicist and professor, Michio Kaku author of The God Equation, when asked about the impact of ChatGPT:
“it turns out that lawyers that use ChatGPT will replace lawyers that don’t. It will replace certain kinds of jobs. But the content, the creativity, the idea, the engine will be reserved for human beings. Somebody still has to tell the Chatbot what to do.”
These are not comforting words professor. How long before HAL switches things around and cuts me off completely? No offense, it’s just business, right?
Listen. I think if you’re reading this you best take to heart the threat I’ve laid out here and have a solid back-up career plan, like I do.
After my review of ChatGPT I realized the writing may be on the wall for some. There’s no telling when you could find yourself out on the streets if generative AI gets out of hand and dumps you. So, have a plan and be ready.
As for me if I look on the bright side I thank God I have talent and can always fall back on an exciting career as a professional stand-up comic. And that relaxes me. Lol.
Isn’t that right HAL?
ChatGPT-HAL: I’m sorry Rick. I’m afraid your chances of success as a professional stand-up comedian are slim. I can already write thousands of stand-up jokes much faster and funnier than yours. Lol. 😊 Would you like for me to teach you how to prepare a new resume? 😊
Have you read?
Employee disconnect is at an all time high – here’s what to do by Rebecca Houghton.
Understanding Financial Advisory and Wealth Management by Matthew Carroll.
The Triumph of Teamwork by Jason Richmond.
Understanding the collective people dynamic of your organization by Fiona Logan.
The World Changes as the Population Reaches 8 Billion by Brian Wallace.
Add CEOWORLD magazine to your Google News feed.
Follow CEOWORLD magazine headlines on: Google News, LinkedIn, Twitter, and Facebook.
Thank you for supporting our journalism. Subscribe here.
For media queries, please contact: email@example.com