Views: 4
Can books on AI help us to comprehend the digital transition, or at least part of it? Many authors would certainly think so, though the question cannot be truly answered, as the impact of some books can only seldom be empirically proven. This is not the case of the Mollick’s book, as it is both theoretical and practical.
The first general purpose AI (mainly in the ChatGPT version) was born at the end of 2022. But even years before, and more actively since then, most researchers started to investigate the numerous AI applications role and place in business, national economies and human life. One of the recent books attracted our attention: below is a short review we’d like to share with our readers.
Interesting approach to “trained” AIs
The “co-intelligence” book is a concise specification of the origin and “formative years” of the digital transition (with the LLMs and AIs, to name a few), the process’s turning points and numerous factors that shaped its trajectory, as well as presenting a view of AI’s possible future. In eight chapters and an epilogue he tracks the AI’s place within the wider digital context over time and describes the essence of the AI’s co-intelligence involvement as a “person”, creative coworker, tutor and coach.
Researchers around the world paid attention to the emerging possibilities to perform a sort of creative and innovative work that were the result of human activity previously. Besides the ChatGPT, some other algorithms appeared, such as e.g. Artificial General Intelligence (AGI), as well as the General Purpose Technologies and Large Language Models (numerous LLMs), to name a few. Although the term “artificial intelligence” was invented in 1956 (!) by John McCarthy of MIT in the US, the latest AI boom started in the 2010s with using machine-learning techniques for data analysis and predictions. With wide introduction of AI algorithms, the digital science moved to more accurate assessment, predictions and even co-intelligence!
As E. Mollick*) notes (the book was finished at the end of 2023) that researchers still strive “to teach AI to understand and generate humanlike writing” by training huge amounts of text from various sources, such as e.g. websites, books and other sources in the digital form. (p.10)
A fruitful year for publications on AI-2024
This year has been an extremely fruitful year for both popular and scientific publications with about eight interesting books on the issue.
Among them was a book by the US Wharton professor Ethan Mollick who throughout his writing and teaching has become one of the most prominent and provocative explainers of AI, focusing on the practical aspects of how these “new tools for thought” can transform societies. However, with some reservations: e.g. contrary to traditional software (which is quite predictable) the AI “fabricates an answer rather than reflecting its own processes… hence, there is no definite guide on how to use AI in your own organisation”. (p.66)
The AI and ChatGPT meant numerous advantages for humans who after millions of years on their own managed to develop a kind of co-intelligence means that could augment, or even replace, human thinking. Thus, Mollick urges the reader to engage with AI as a “co-intelligent co-worker, co-teacher and coach” at the same time. He assesses its profound impact on business and education, using dozens of real-time examples of AI in action. The book explicitly shows what it means to think and work together with “smart machines”, and why it is imperative that people have to master these digital skills.
Mollick challenges us to utilize AI’s enormous power without losing our identity, to learn from it without being misled, and to harness its gifts to create a better human future. Wide ranging, hugely thought-provoking, optimistic and lucid, the book reveals the promise and power of emerging new AI-era.
*) General reference to: Mollick E. Co-Intelligence: Living and Working with AI (2024). WH Allen, Penguin Random House, UK. – 234 pp. ISBN-13-978-0753560778.
Note. Additional recent books on the AI issues (all published in 2024)
-Voskanyan T. AI Guide for Beginners: How to Use AI Prompts & Master Artificial Intelligence in 4 practical days (21 Days to Make Money with AI). 2024. -110 pp. ISBN-13-979-8346299585.
-Lembo G. The Complete Beginner’s Guide to AI: Your Easy Guide to Artificial Intelligence, Real World Applications, Ethical Issues, and Future Innovations. 2024. -124 pp. ISBN-13- 979-8339377900.
-Understanding Generative AI for Business Leaders: Demystifying Strategic Advantage, Ethical Deployment and Practical Integration for Success (Artificial Intelligence for Everyone series). 2024. -186 pp. ISBN-13- 979-8894432090.
-Seth N. Mastering the Data Paradox: Key to Winning in the AI Age. 2024. -584 pp. ISBN-13-978-0143468035.
-Harari Y. N. Nexus: A Brief History of Information Networks from the Stone Age to AI. 2024. -528 pp. ISBN-13-978-0593734223.
-Shive E. Understanding Generative AI for Beginners: Fundamentals of Artificial Intelligence, Practical Applications of ChatGPT, and Simple Methods to Build Foundational Knowledge for Elevating Tech Projects. 2024. -178 pp. ISBN-13-979-8328548601.
-Sarr M. A. The AI Entrepreneur: How to Make Money with AI: From Idea to Launch — Build, Fund, Market, and Scale Your AI Business in 90 Days or Less. 2024. ISBN-13-979-8343087956.
-Grant R. AI for Business: The Beginner’s Fast Track to ChatGPT for Productivity, Profit, and Growth (2 books in 1). 2024. -385 pp. ISBN-13-978-1962079129.
Source: https://www.amazon.com/dp/1962079120/ref=sspa_dk_detail_6?psc=1&pd_rd_i=1962079120&pd_rd_w=U8QbS&content-id=amzn1.sym.386c274b-4bfe-4421-9052-a1a56db557ab&pf_rd_p=386c274b-4bfe-4421-9052-a1a56db557ab&pf_rd_r=AK00H6BASQF605069DFQ&pd_rd_wg=AvnH4&pd_rd_r=98584ad3-7a31-467d-9863-cd21ed0c82a9&s=books&sp_csd=d2lkZ2V0TmFtZT1zcF9kZXRhaWxfdGhlbWF0aWM
Some other interesting ideas…
In the chapter on AI as a person, the author provides an example of “current assessment of the LLMs’ intelligence, referring to a famous paper by a team of Microsoft researchers (March 2023) which claimed that GPT-4 (the latest and most powerful language model developed by OpenAI) “exhibited signs of general intelligence” (p. 86)
“Sparks of Artificial General Intelligence: Early Experiments with GPT-4” (2023), Cornell University, in:
https://doi.org/10.48550/arXiv.2303.12712
However, notes Mollick, we have to remember that “AI is not human, but often works in the way that we expect humans to act”, and finally (refereeing to Bing image creator: https://www.bing.com/images/create) that any judgment (sentient) “is not a fixed or static state, but a dynamic and evolving process” (p. 92)
Finally, the author comes to a conclusion full of promises and regrets: on one side postulates that “AI is a mirror, reflecting back at us our best and worst qualities”; i.e. the people will finally decide on AI implications in future. On another side, the AI is a “co-intelligence, not a mind of its own; humans are far from obsolete, at least for now”. What a beautiful conclusion…
European picture
In line with the Mollick’s ideas, the EU-wide scene in facilitating AIs is developing quickly. Thus, under pressure from competitors like DeepSeek and Meta’s Llama 3, OpenAI said it is working on releasing a new open-weight large language model in the coming months. Then , MetaAI will be launched across EU-27 states some months after the company faced regulators’ questions for the use of personal data to train its AI models. The European Commission is awaiting a risk assessment by Meta to check whether the tech giant’s new AI chat function meets obligations under the EU online platform rules, said recently the spokesperson for the Commission, adding that “certain aspects” of MetaAI fall within the scope of the Digital Services Act (DSA), which sets out standards for safety and transparency in the EU-27. Thus, under the DSA, a company needs to submit risk assessments to the Commission annually but also in advance of deploying new functions.
“Once we have received the risk assessment, we will carefully analyse all relevant documents to ensure compliance with the DSA and that the feature does not present any undue risks within the European Union,” the spokesperson said.
Meta announced lately that it will roll out MetaAI across the EU; it was rolled out in the US in September 2023, followed by India in June 2024, and in the UK in October 2024. Its plan to continue expansion in Europe was halted last summer due to “regulatory unpredictability”: the Irish Data Protection Commission told the company to postpone its plan because it planned to use data of adult users of Facebook and Instagram to train its large language models (LLMs).
“It’s taken longer than we would have liked to get our AI technology into the hands of people in Europe as we continue to navigate its complex regulatory system – but we’re glad we’re finally here,” Meta said in a blog post. A spokesperson for the company said in a statement to Euronews that Meta has been “fully transparent with the European Commission on the launch of Meta AI”.
“We’re confident in our compliance and remain in ongoing dialogue with the Commission,” the statement added.
The company – including its CEO Mark Zuckerberg and global policy chief Joel Kaplan – have been critical of Europe’s regulatory action against US tech companies, even more so after the new US administration of Republican President Donald Trump took office in January.
General source: https://www.euronews.com/next/2025/03/24/metas-ai-tool-still-not-cleared-for-use-by-eu-commission?utm_source=newsletter&utm_campaign=feed_next_ia&utm_medium=referral&insEmail=1&insNltCmpId=2012&insNltSldt=10080&insPnName=euronewsfr&isIns=1&isInsNltCmp=1