Macrotrends Driving Business Transformation...

... or why managers and leaders should explore new technologies. Also in this issue: Llama 2 and Apple Vision Pro.

This is MadeMeThink.xyz – A weekly publication with thoughts on disruptive Technologies (e.g. AI | Web3 | Metaverse), their Applications, Opportunities, and Impact. Written by Prof. Thomas Metzler, Ph.D.

This MadeMeThink…

Macrotrends driving business transformation…

The World Economic Forum has done a study on "The Future of Jobs". It's a really extensive read. I flipped through it and got stuck on the graphic shown below with the following description: “Trends ranked by share of organizations surveyed that identified this trend as likely or increasingly likely to drive transformation in their organization”.

Not surprisingly for me, rank number 1 is “increased adoption of new and frontier technologies”. In my opinion, managers in companies should be up to date with new technologies. The point is not to understand new technologies in depth, but to be able to assess their potential for their own organization. I can't tell you how many times I've met executives who pushed new trends or technologies in the company out of sheer lack of knowledge, because they had no idea about the field, but they were told it was important (how many social media presences or apps fall into this category?). Managers should not rely purely on consultants or agencies, as they often primarily pursue their own economic interests. Managers must therefore be familiar with the most important future technologies in order to be able to assess whether their use will bring value in their context.

Some background to the study:

“The Future of Jobs Survey was conducted in late 2022 and early 2023 bringing together the perspective of 803 companies – collectively employing more than 11.3 million workers – across 27 industry clusters and 45 economies from all world regions. The Survey covers questions of macrotrends and technology trends, their impact on jobs, their impact on skills, and the workforce transformation strategies businesses plan to use” (WEF, 2023, Future of Jobs)

Source: WEF, 2023, Future of Jobs

Not from this study, but a news item that fits the Future of Jobs subject and caught my attention: “IBM has put a freeze on hiring for about 7,800 jobs it thinks will be automated in the near future” (Bloomberg, 2023). In the last issue of my newsletter, I showed a figure from a study that shed light on which jobs will be eliminated first.

Meta’s Llama 2

Interesting move from Meta. Their large language model Llama 2 is free for research and commercial use, I like. There are already several chatbots that use Llama2. You can try them for free here or here.

“Meta is going all in on open-source AI. The company is today unveiling LLaMA 2, its first large language model that’s available for anyone to use—for free (…) The idea, Al-Dahle says, is that by releasing the model into the wild and letting developers and companies tinker with it, Meta will learn important lessons about how to make its models safer, less biased, and more efficient.” (MIT Technology Review, 2023)

If you want to do a deep dive and look "under the hood" of Llama 2, I recommend you take a look at the accompanying published research paper.

The capabilities of LLMs are remarkable considering the seemingly straightforward nature of the training methodology. Auto-regressive transformers are pretrained on an extensive corpus of self-supervised data, followed by alignment with human preferences via techniques such as Reinforcement Learning with Human Feedback (RLHF). Although the training methodology is simple, high computational requirements have limited the development of LLMs to a few players” (Llama 2 research paper).

Below you see an illustration from the research paper on how Llama2 was trained:

“Training of Llama 2-Chat: This process begins with the pretraining of Llama 2 using publicly available online sources. Following this, we create an initial version of Llama 2-Chat through the application of supervised fine-tuning. Subsequently, the model is iteratively refined using Reinforcement Learning with Human Feedback (RLHF) methodologies, specifically through rejection sampling and Proximal Policy Optimization (PPO). Throughout the RLHF stage, the accumulation of iterative reward modeling data in parallel with model enhancements is crucial to ensure the reward models remain within distribution.” (Read the research paper)

Apple Vision Pro

Haha, that escalated quickly 😅 Jim Cramer (2m followers on Twitter, runs the CNBC Investment Club) got the chance to try the Apple Vision Pro and wrote a comment about his experience:

“I was skeptical. What’s the point? I have two eyes. Why do I need them augmented? Why would I want myself to be cut off from whoever I was with to put this helmet on? Then I did it and realized that nothing would ever be the same. It’s that much of a game-changer. Here are five reasons why” (for his five reasons keep reading on Jim Cramer, CNBC, 2023).

Personal update…

Here's what else has been going on for me since the last MadeMeThink issue…

Keynote Surgence Web3 Community

I gave an Online Keynote for the Surgence community on Lean Startup & Business Model Innovation. Surgence HQ is in Sidney, members are spread all over the world, yet we shared thoughts and perspectives easily with each other via a Discord video call. That's one of the things that excites me about Web3, the exchange of thoughts with pioneers all over the globe.

Online Keynote on Lean Startup and Business Model Innovation

Holiday Vibes at Lake Constance

For me, summer is one of the few times I really take a vacation. My vacation has just started this week, I am currently enjoying the time at Lake Constance.

Doing some Offline Time at Lake Constance (in Austria/ Vorarlberg, where I live).

Disclaimer: The thoughts published in this publication are my personal opinions and should not be viewed as investment advice. I am not a financial expert. My specialty is entrepreneurship, marketing & innovation. Readers should always do their own research.