How long...really...until ChatGPT changes everything?

How long...really...until ChatGPT changes everything?

The NY Times in an article last week said that there will be a lag from the introduction of Generative AI platforms (like ChatGPT, BERT, Dolly) and the time where productivity begins to increase, as happened in past technology revolutions. The lag this time may be shorter than history has shown us for a few reasons:

  1. Electrification: Factories had to be rebuilt to create productivity gains when the steam engine was replaced by the electric motor. It took a generation to realize significant gain because the physical impediments and sunk cost of existing technology made it hard for humans to accept the change and reimagine the factory.
  2. Computerization: When computerization happened, we once again had sunk cost in secretaries, accountants, copy writers, paper-to-digital transcription and lots of physical cost for machines and storage. It took a generation to redefine education, hiring and work processes to take advantage of the digital capture and use of knowledge and process. In many ways, it's still being worked out today, fifty ears after the first PCs showed up.
  3. Generative AI: Everything exists today to capitalize quickly on the digital storage, the machines and the kinds of problems that can simply be worked out far more quickly with a minimum amount of human retraining. A new layer of APIs can look at vast amounts of data to very quickly figure out answers to very complex tasks. It isn't replacing human work...it is better than any individual human's work.

Generative AI is the payoff for all of the investment in digitization over the past two generations. There's not a sunk cost problem because there isn't a wholesale replacement of physical machines, storage and workspaces. Each of these new applications of Generative AI is simply a way to access a vast amount of structured and unstructured information as it lies through a "large language model." Enormously large data sets of all kinds give up intelligence in ways very different from the data sitting under any one application, like Salesforce, SAP, in every proposal you've written in the past x years, regardless of its form factor. This is intelligence breaking free from where and how data is used.

I tested GPT-4 to see what it could give me in the following tasks:

  • Writing a termination letter for an employee that complies with regulations, sets an appropriate tone, and provides solid rationale. The longest element of the task is coming up with what needs to be said, not the way to say it. A few points and the outcome was generated in minutes where it was previously hours of creation and perhaps days of review.
  • Creating a proposal based on a set of use cases that prices appropriately for buyer and seller regardless of product, pricing, terms and other complexities. It was done in minutes with just a few refinements.
  • Producing a job description outlining the most important skills of a B2B software sales person. In my test, I kept refining the choices from B2B software, to CRM software, to large enterprise sales. Each time it refined in exactly the way I would expect and produced excellent content.
  • Drafting an inoffensive note to a spouse asking them to do the dishes more often. Nailed it.

We're on the cusp of fast-moving change that isn't media hype and will change the nature of work. The winners will be those who harness it quickly and wisely and the losers will be those who don't. Get moving!

Applying the technology in this article:

For fun, I asked ChatGPT, "write an article about how quickly Generative AI will be adopted compared to electrification and computerization." The answer was as follows:

"Generative AI is a powerful technology that is poised to transform a wide range of industries, from healthcare and finance to art and music. While the potential applications of generative AI are vast, the speed at which it will be adopted remains a matter of debate. In this article, we will compare the adoption rates of generative AI to those of two other transformative technologies: electrification and computerization.

Electrification refers to the widespread adoption of electrical power as a means of generating energy. This process began in the late 19th century and took several decades to become fully established. While the benefits of electrification were clear, including improved productivity, increased safety, and greater convenience, it took time for the necessary infrastructure to be put in place and for people to become familiar with the new technology.

Similarly, computerization, or the widespread adoption of computers and digital technology, began in the mid-20th century and has continued to accelerate over the past several decades. The benefits of computerization, including increased efficiency, greater access to information, and improved communication, have been widely recognized. However, the process of adoption has been uneven, with some industries and individuals embracing the technology more quickly than others.

In comparison to these two technologies, the adoption of generative AI is likely to be both faster and more widespread. One reason for this is the speed at which technology is now advancing. Compared to the development of electricity and computers, which took decades to become fully established, generative AI is developing at an exponential rate, with new breakthroughs and innovations emerging on a regular basis.

Another reason for the faster adoption of generative AI is the potential benefits it offers to businesses and individuals. Unlike electricity and computers, which were primarily adopted for their productivity and efficiency benefits, generative AI has the potential to transform a wide range of industries in more fundamental ways. For example, in healthcare, generative AI can help doctors make more accurate diagnoses and develop personalized treatment plans, while in the arts, it can help artists create new works that were previously unimaginable.

In addition, the adoption of generative AI is likely to be accelerated by the growing awareness and understanding of the technology. While there is still a significant knowledge gap between technical experts and the general public, the pace of education and awareness is likely to be much faster than with previous technologies. As more individuals and organizations become familiar with the capabilities and limitations of generative AI, they are likely to become more receptive to adopting it.

Of course, there are still challenges that need to be addressed before generative AI can be fully adopted. These include regulatory and ethical considerations, as well as concerns around data privacy and security. However, these challenges are not insurmountable and can be addressed with appropriate safeguards and oversight.

In conclusion, while the adoption rates of electrification and computerization took several decades, the adoption of generative AI is likely to be faster and more widespread. This is due to the rapid pace of technological advancement, the potential benefits the technology offers, and the growing awareness and understanding of the technology. While there are still challenges that need to be addressed, the future of generative AI looks bright, and we can expect to see it become an increasingly important part of our lives in the years to come."

No alt text provided for this image
Usama Tamimi

Building your affordable remote offshore team in just 1 week! ???? Connecting ??US and European companies/startups with top offshore affordable talent.??

1 年

Chris, thanks for sharing!

David Williams

Open Source AI Leadership

1 年

You may enjoy this Lex Fridman podcast with Eliezer Yudkowsky, Chris. The "death of humanity" messaging is a bit alarmist in it, but I think there are some solid points on how difficult it will be to navigate the phase transition from current LLMs to what will be engendered next. https://lexfridman.com/eliezer-yudkowsky/

David Whitehouse, Ph.D.

Retired - ex Databricks

1 年

So true! It's hard to fathom how quickly this will become part of our lives

The productivity gains will come sooner than people realize!

要查看或添加评论,请登录

社区洞察

其他会员也浏览了