Hello! While we haven't publicly communicated much on the case, Lokad has been very active on the LLM front over the last couple of months. We have also an interview with Rinat Adbullin, coming up on Lokad TV, discussing more broadly LLMs for enterprises.

LLMs are surprisingly powerful, but they have their own limitations. Future breakthrough may happen, but chances are that whatever lift some of those limitations, may be something quite unlike the LLMs we have today.

The first prime limitation is that LLMs don't learn anything after the initial training (in GPT, the 'P' stands for 'pretrained'). They just perform text completions, think of it as a 1D time-series forecast where values have been replaced by words (tokens actually, aka sub-words). There are techniques to cope - somehow - with this limitation, but none of them is even close to be as good as the original LLM.

The second prime limitation is that LLMs deal with text only (possibly images too with multi-modal variants, but images are most irrelevant to supply chain purposes). Thus, LLMs cannot directly crunch transactional data, which represents more than +90% of the relevant information for a given supply chain.

Finally, it is a mistake to look at the supply chain of the future, powered by LLMs, as an extension of the present-day practices. Just like eCommerce companies have very little in common with mail-order companies that appeared in the 19th century; the same will - most likely - be true for those future practices.

Thank you for the detailed insights!. It's always enlightening to hear from experts like Lokad. While I understand Lokad's active involvement and experimentation with LLMs, I'd like to share my perspective based on your points and my own observations.

Firstly, the limitations you mentioned regarding LLMs, especially their inability to learn post their initial training, is indeed a significant challenge. Their textual (and sometimes image) processing capabilities, though remarkable, may not suffice for the intricate nuances of supply chain transactional data. This is especially true when considering that such data comprises over 90% of pertinent supply chain information.

However, I envision generative AI working in tandem with supply chain teams, not replacing them. The role of a learning agent (at each node) could be used to assist these teams, enabling them to capture a more comprehensive representation of unconstrained demand and thereby enriching their baseline models. While the potential of AI in supply chains is vast, solely relying on this technology for business practices might be too early, given its experimental nature.

In the future, I believe the challenge won't be about replacing current practices with LLM-powered processes but merging both to create a hybrid model where technology complements human expertise. Just as eCommerce companies evolved from but differ vastly from mail-order companies of the 19th century, our future supply chain practices will likely be an evolution, not a replacement, of current methods.

I am not overly convinced by the idea of 'agents' when it comes to system-wide optimization. Indeed, the optimization of a supply chain must be done at the system level. What is 'best' for a node (a site, a SKU, etc) is not the what is best for the whole. The 'agent' paradigm is certainly relevant for modeling purposes, but for optimization, I am not so sure.

Concerning evolution vs revolution, see 'Incrementalism is the bane of supply chains', https://www.lokad.com/blog/2021/9/20/incrementalism-is-the-bane-of-supply-chains/