Supply Chain Matters provides a technology focused editorial perspective focused on the existing interest and momentum of advanced generative AI technologies and their application in business processes.
Background
The continuous news feeds and various perspectives surrounding deep learning generative artificial intelligence (AI) or termed Chat GPT systems are endless. Depending on which stakeholders are generating such perspectives, the implications vary from the next renaissance of advanced technology to a threat to jobs, livelihoods and to humanity itself.
All are pertinent.
Wikipedia describes Chat GPT technology: “as an artificial intelligence chatbot developed by OpenAI and released in November 2022. It is built on top of OpenAI’s GPT-3.5 and GPT-4 foundational large language models (LLMs) and has been fine-tuned (an approach to transfer learning) using both supervised and reinforcement learning techniques.”
Reportedly, the function of the chatbot can be to mimic human conversation with responses that could report answers to a specific problem, the creation of a report or an opinion. The technology can also be utilized to emulate certain software systems including generating actual code, graphic images or debugging of systems.
Microsoft, after an initial $1 billion investment in 2019, invested an additional $10 billion in OpenAI in January of this year. The enterprise tech provider has plans to embed ChatGPT technology in both its various Azure Cloud based systems along with the company’s Bing enterprise search engine. Micosoft CEO Satya Nadella has equated this technology towards reshaping just about every software category. This, in-turn, has initiated a generative AI arms race among Alphabet (Google), Amazon AWS and other Cloud and software tech providers.
Both OpenAI and Microsoft are reportedly taking a “cautious approach” on how this technology is going to generally be deployed. They readily acknowledge that this technology remains prone to inaccuracies because the models are still learning. They further cannot currently discern what is copyrighted or intellectual property protected content, leaving that decision to the user who initiated the request.
Many supply chain technology providers have already leveraged forms of existing AI-enabled planning and decision-making in supply chain business processes. Thus, the opportunity to leverage or utilize generative AI models has captured competitive interest levels among technology providers. The Wall Street Journal has reported that the likes of Blue Yonder, SAP SE and Manhattan Associates have active research programs underway. The Wall Street Journal reported in April that retailer Walmart is utilizing an AI chatbot developed by Pactum AI to automate indirect materials supplier negotiations based on the retailer’s parameters for budget and supply needs. Reportedly the software communicates directly with human sellers for items such as shopping carts, reducing negotiating times to days instead of weeks.
Our belief is that interest levels among tech companies are even broader. That is because overall market growth rates for Cloud based infrastructure and software revenues have been on a decline. Generative AI is a basis of new hype for technology solution, but the question remains when will the technology be both mature and safe enough.
Statements and Actions of Concern
From a Supply Chain Matters perspective, this is our first commentary on this technology.
We initially wanted to gain a better understanding of what is behind this technology in terms of approaches, backers and supply chain business process solutions perspectives. We remain in this market educational effort. Thus, this initial commentary will delve into what we believe are the most obvious concerns and for what our readers should be watchful.
Voices regarding a perceived new AI arms race without adequate protections in the use and deployment of generative AI are many at this point.
In late March, an appeal from a number of technology executives and AI researchers called for a six month moratorium to allow the industry the time to establish safety standards and/or guardrails. The signers of that appeal included Apple Co-Founder Steve Wozniak, Stability AI CEO Emad Mostaque, Tesla and Twitter Senior Executive Elon Musk, and the two co-founders for the Center of Humane Technology, Tristan Harris and Aza Raskin.
The Economist published an editorial indicating that: A battle royal is brewing over copyright and AI, (Paid subscription required) as lawsuits have begun over extracting content or databases without permission or compensation. Specifically noted: “It is a legal minefield with the implications that extend beyond the creative industries to any business where machine-learning plays a role, such as self-driving cars, medical diagnostics, factory robotics and insurance risk management.” Getty Images has already filed a lawsuit accusing start-up Stability AI of infringing on copyright violations among a reported millions of images. Others are likely to follow.
In April, the European Data Protection Board (EDBP) announced plans to foster a dedicated task force to assess ChatGPT after a number of privacy entities raised concerns as to whether the technology is compliant with the EU’s General Data Protection Regulations (GDPR). Italy’s data privacy regulator had already imposed a temporary ban on the technology over alleged data privacy concerns. European Union lawmakers are reportedly in the process of moving toward passage on new policies designed to better regulate generative AI, principally to protect the rights of citizens.
Earlier this month, The Biden Administration indicated that while advanced AI could be a major opportunity, there are inherent dangers of the technology to public safety, privacy and democracy, if this technology were to be deployed by bad actors. A termed constructive discussion and exchange of ideas was held with major industry executives.
Geoffrey Hinton, considered to be one of original founders of AI technology concepts announced his departure from Google parent company Alphabet, so that he could add his voice to a growing number of tech executives about the potential threats that could be posed if AI systems achieved greater intelligence then that of humans. Hinton stated flatly his belief that advanced AI could pose a “more urgent” challenge than that of climate change.
The Wall Street Journal’s CIO Journal has featured recent reports of industry CIO’s placing temporary bans on internal use of generative AI tools because the current rapid pace of development has not allowed adequate time to rationalize the implications as well as the guardrails of use.
Specific restrictions among companies include J.P. Morgan’s reported restriction on the use of generative AI technology among staff. Amazon reportedly warned its employees not to upload any deemed confidential information into such systems. Google itself reportedly has a specific internal policy with its own Bard generative AI application that restricts the company’s employees from uploading any internal documents that would be sensitive to the business.
In late April, Samsung was forced to temporarily restrict the use of generative AI tools after the tech giant discovered misuse among employees leading to a reported three separate incidents of the leaking of deemed sensitive business data.
Tech Industry Start-up Replays
What has especially caught our interest was a report by The Wall Street Journal published earlier this month, No Business Plan? No Problem. ChatGPT Spawns an Investor Gold Rush in AI. (Paid subscription or metered view) The essence of the report is that while most of Silicon Valley’s venture-capital ecosystem remains in the doldrums, investors this year have been pouring funds into unproven generative AI start-ups with PitchBook predictions of a potential $43 billion market size this year this year alone.
These new AI researcher based start-ups are reportedly garnering hundreds of millions in initial funding rounds with hyped-up valuations without so much as a business plan.
Initial Takeaways for Readers
The warnings and statements of concern speak for themselves.
Businesses will need to heed such warnings in evaluation of the timetables and uses of this technology.
From our lens, the tech industry is once again in a cycle of generating as much hyped news regarding the game changing potential of this technology. This is not productive, nor helpful, because market education is lacking, especially toward the implications.
Tech marketing teams need to tone down the hype and instead focus on what specific business and societal needs that generative AI can and will address.
Many in our world community continue to be impacted by social, geopolitical and demographic changes occurring across the globe and on the perceived harms divided and contemptuous politics have brought forward to families and communities.
It is meaningful that government and tech leaders raise the issue of proper safeguards since bad actors with more sophisticated cyberattack or other data and information manipulation tools can bring even greater harm to societies.
Some profess that the use of this new advanced AI technology will be a game changer, and at the same time, new and far different jobs will be generated. That remains to be seen or proven in evidence and practice.
MIT Professor Yossi Sheffi has published a new book, The Magic Conveyor Belt, Supply Chains, AI and the Future of Work. This Editor is in the process of absorbing the many observations presented, including the generation of new and different supply chain roles. At the same time, Professor Sheffi provides thoughts on the vital links and context that humans will continue to provide for supply chain decision-making needs. We will feature added commentary on this text at a later date.
Some enterprise tech companies, a good example being SAP SE, are openly acknowledging a business and social responsibility approach to deep learning AI technologies. That includes respecting the privacy of data, along with transparency as to how advanced AI technology learns and makes inferences leading to specific recommendations. More of such pledges and actions should be demanded by businesses of their information technology providers.
In open disclosure, we may have our own biases related to Chat GP. This blog prides itself on our independent thought leadership content that have been a part of this blog since our founding in 2018. That stated, we also protect all of our published content with a registered U.S. copyright.
Generative AI requires responsibilities and actions from multiple dimensions, businesses, AI and enterprise tech companies, and governmental oversight of personal protections. Time is needed for alignments and added education to occur and marketing hype will not alter that reality.
Bob Ferrari
© Copyright 2023, The Ferrari Consulting and Research Group and the Supply Chain Matters® blog. All rights reserved.
