OpenAI's AI Supremacy Strategy
Also: Ex-Google CEO: "Pull the Plug!"
Dear Friends,
this is a weekend special edition of The Augmented Manager Newsletter, for two reasons:
I'm on sick leave, so I have more time... but this is not that important
Most importantly: OpenAI released its policy recommendation to secure AI supremacy for the U.S.. That made me think - and write this edition.
In this Issue
Battle for AI Supremacy: OpenAI's Strategy 🔥
Ex-Google CEO Eric Schmidt: "Pull the Plug!"
AI Tools We Love: NotebookLM 🛠️
Culture Corner: Digital Twin in Vatican
AI Art: Prince in the Purple Rain
Prompting Like a Pro: Midjourney Image Generation
Quote for You: Rob Garf, Salesforce💡
Battle for AI Supremacy
In the rapidly evolving world of artificial intelligence, OpenAI has positioned itself as a leader by developing a multifaceted strategy that emphasizes innovation, collaboration, and ethical considerations. This article examines OpenAI's strategic development, highlighting its initiatives to compete globally, particularly against China, and the implications for businesses navigating this transformative landscape.
OpenAI's Strategic Evolution: From a non-profit company to a 157 billion USD enterprise
Founded in 2015 by prominent tech leaders including Elon Musk and Sam Altman, OpenAI was established with a mission to ensure that artificial general intelligence (AGI) benefits all of humanity.
"OpenAI is a nonprofit artificial intelligence research company. Our goal is to advance digital intelligence in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate financial return. Since our research is free from financial obligations, we can better focus on a positive human impact."
The founders recognized the transformative potential of AGI while also acknowledging the associated risks. Thus, they committed to developing AI technologies transparently and collaboratively, aligning them with human values and interests.
OpenAI's initial model was non-profit, but in 2019, it transitioned to a "capped-profit" structure, as the company established a "for-profit arm". That year, the original nonprofit established a for-profit subsidiary to attract the substantial investments required to develop cutting-edge AI. However, it took an unprecedented step in Silicon Valley by limiting investor profits. Investors could earn up to 100 times their initial investment, but any returns beyond that would go back to the nonprofit. The nonprofit could then use these funds to serve the public good, such as financing a universal basic income program to help people adapt to job displacement caused by automation.
This hybrid model allowed for attracting funding while ensuring that the majority of profits are directed towards its overarching mission of benefiting humanity. This shift has enabled OpenAI to scale its operations significantly while maintaining a focus on responsible development.
Innovative Product Development
From 2022 onwards, OpenAI has made significant strides in product development. The launch of GPT-4 and subsequent models has revolutionized natural language processing (NLP), enabling applications across various sectors such as customer service, content creation, and education. These advancements have been complemented by multimodal capabilities, allowing models to process both text and images. Key products include:
GPT-4: A powerful language model capable of generating human-like text and engaging in complex conversations. The GPT-4o and o1 go even beyond, GPT-4o acts as multimodal LLM covering image and video, o1 preview has extreme advanced reasoning capabilities.
Advanced Search and AI Agents: A few days ago, the company introduced ChatGPT Search, and in January 2025 they will release "The OPERATOR", a very capable AI Agent, a competitor of Microsoft's Magentic-ONE and Anthropic's Claude with Computer Use.
DALL·E: An AI system that creates images from textual descriptions, showcasing the potential for creativity in AI applications.
Codex: A model designed for understanding and generating code, streamlining software development processes.
These innovations not only enhance OpenAI's product offerings but also provide businesses with tools to improve efficiency and drive innovation.
Strategic Partnerships
OpenAI's partnerships have been crucial in expanding its reach and capabilities. The ongoing collaboration with Microsoft is particularly noteworthy. This partnership has facilitated access to Azure's cloud infrastructure, enabling OpenAI to deploy its models at scale while benefiting from Microsoft's extensive resources. Additionally, collaborations with academic institutions and other tech firms have fostered an ecosystem of innovation. For instance:
Partnerships with universities: These collaborations focus on research initiatives that explore ethical AI development and safety measures.
Alliances with industry leaders: By working with companies across various sectors, OpenAI aims to tailor its technologies for specific applications, enhancing their practical utility.
The remarkable history of OpenAI until recently is shown on this chart below:
Source: Coffeespace.com
Sam Altman's Turnaround
In recent years, OpenAI gradually shifted its priorities away from safety, focusing instead on rapidly commercializing its products. By 2023, the nonprofit board had grown so distrustful of Altman that they attempted to remove him. However, he quickly regained control, leveraging his ties with Microsoft and installing a new board more aligned with his interests. Earlier this year, OpenAI’s safety team fell apart as staff members, disillusioned with Altman’s leadership, resigned.
In September 2024, Altman has taken the ultimate step to solidify his authority: he has effectively stripped the board of its power. While the board will technically remain, it will no longer have any meaningful influence.
The original nonprofit has been stripped of its power, and its mission has been redefined to prioritize profit.
Because profit might be exactly what Altman believes the company urgently requires. Despite a boldly optimistic blog post, where he claimed AI would contribute to “fixing the climate, establishing a space colony, and uncovering all of physics,” OpenAI faces significant challenges. It has struggled to identify a sustainable path to financial success for its models, which demand investments in the hundreds of millions — if not billions. Transitioning the business to a for-profit model could make it more appealing to investors.
However, this move has raised eyebrows — including those of Elon Musk — with critics asking: Is this even legal?
Eliminating the profit cap could redirect substantial sums of money — potentially billions in future revenue — from the nonprofit to private investors. Given that the nonprofit is intended to serve the public interest, this shift could effectively divert billions away from ordinary people. To many, this feels alarmingly close to theft.
“If OpenAI were to retroactively remove profit caps from investments, it would effectively transfer billions in value from a nonprofit to for-profit investors,” said Jacob Hilton, a former OpenAI employee who joined prior to its transition from a nonprofit to a capped-profit structure. “Unless the nonprofit were properly compensated, this would be a blatant money grab. In my view, such a move would violate OpenAI’s charter, which states that its primary duty is to humanity. I fail to see how the law could allow this.”
Due to OpenAI’s unique structure, the legality of such a change is murky. This ambiguity might be precisely what the company is relying on.
When asked for comment, OpenAI referred to its statement in Bloomberg, where a spokesperson asserted that the company remains “focused on building AI that benefits everyone” and emphasized that “the nonprofit is core to our mission and will continue to exist.”
The U.S. Presidential Elections and OpenAI's Move into Politics
After the US presidential elections and Trump's win, the tech sector surged, and OpenAI strengthened its dominance on the AI market. The time has come to step up and make the first steps toward politics.
Over the past year, numerous local governments have made significant strides in adopting emerging artificial intelligence (AI) technologies, applying them to streamline governmental tasks and establishing guidelines for their use.
OpenAI, the once non-profit, now 157 billion dollar enterprise, envisions an even greater potential for this technology. The company has unveiled a comprehensive blueprint for building AI infrastructure in the U.S., emphasizing the importance of federal, state, and local community collaboration.
Published on November 13, the blueprint outlines five ambitious pillars aimed at transforming American cities and counties. In an interview with the Center for Strategic & International Studies (CSIS), OpenAI's Vice President of Global Affairs, Chris Lehane, compared these pillars to major U.S. infrastructure initiatives like the National Highway Act of 1956 and the Telecommunications Act of 1996. "We need to start to think big and act big in the same way that we have done in the past," Lehane stated. Watch the full interview here:
OpenAI's five pillars for building AI infrastructure
Development of AI Economic Zones: This pillar proposes federal and state initiatives to expedite permitting and approvals for wind farms, solar arrays, and nuclear reactors dedicated to powering AI infrastructure. To incentivize local communities, a portion of the computational resources generated by these data centers would be allocated for public use, supporting public universities and fostering developer ecosystems.
Passing of a "National Transmission Highway Act": To ensure nationwide access to AI technology, this pillar advocates for the creation of a nationwide "AI highway." This would involve expanding transmission lines, fiber connectivity, natural gas pipelines, and establishing a framework for wireless AI connectivity. Such an act would address the planning, permitting, and funding necessary to build a transmission grid optimized for AI commerce and deployment.
Securing Government Backstops for High-Value AI Public Works: Recognizing the need for substantial investment, this pillar suggests that the government encourage private investors to fund high-cost energy infrastructure projects by committing to purchase energy and implementing measures to reduce credit risk. These privately funded projects would serve the public as strategic national assets. Additionally, there is an emphasis on funding training programs for workers involved in building and servicing these facilities.
Developing a "North American Compact" for AI: This pillar recommends that the U.S. collaborate with allies to support its AI vision and counter emerging AI infrastructure developments in other countries. Such a compact could enhance the resilience of supply lines and strengthen international partnerships in AI development.
Reinvigorating the Nation's Nuclear Power Sector: OpenAI views nuclear power as essential for powering future AI technologies. This pillar calls for strengthening and rebuilding the country's nuclear power capabilities. Citing the U.S. Navy's 75-year history of incorporating nuclear power into submarines, Lehane suggests that similar approaches could be applied domestically to provide the necessary energy for AI infrastructure.
OpenAI's bold move immediately positioned the company within the political arena. Just a few days ago, Elon Musk, the founder of OpenAI and CEO of Tesla, was reportedly offered a formal government position. These developments highlight how politics, power, and tech giants are becoming increasingly intertwined in shaping a new AI-driven global order—hopefully before China or other non-democratic nations can do so.
The new arms race for AI supremacy is accelerating. Powerful AI systems are the Holy Grail of humanity nowadays. Experts have different opinions when will we reach it.
AGI Timeline Estimates by 1700 experts
Source: AI Multiple Research
As you can see, there are a lot of experts who believe, we will reach very powerful AI systems in the upcoming 10-20 years. Actually, the industry leaders incl. Sam Altman think, it might be only a 'few thousand days' away. Those AI systems will not annihilate the complete job market, they will also not solve all of our problems incl. climate change and increasing inequality. But they might be very capable of causing immense harm to humanity when getting into the wrong hands. And that we must prevent.
Takeaways for Managers and Leaders
As AI continues to reshape industries, understanding OpenAI’s strategy and recent events around it offers valuable lessons for business leaders. Here are the critical takeaways:
AI as a Non-Negotiable Organizations must embrace AI to enhance efficiency and drive innovation. Leveraging tools like ChatGPT, large language models (LLMs), and generative AI solutions can revolutionize customer engagement by enabling personalized interactions at scale. As AI becomes ubiquitous, it will emerge as the most critical strategic resource of the coming years.
Forming Strategic Tech Partnerships Collaborations with tech companies are key to accessing cutting-edge technologies and expertise. Such alliances can accelerate digital transformation initiatives and position businesses as early adopters of innovative solutions.
Navigating Politics and Regulation As companies grow and AI development accelerates, engagement with politics becomes almost inevitable. The question is not if but how proactively these organizations should navigate this landscape. For the average companies, staying informed about regulatory changes, such as the EU’s AI Act and forthcoming U.S. regulations, will be essential for compliance and seizing new opportunities.
Securing AI Infrastructure Nations and international alliances are prioritizing the development and protection of critical infrastructure needed to operate AI models. This includes building energy grids, nuclear power plants, and massive data centers. Businesses must align with these efforts to remain competitive.
Cybersecurity as a Cornerstone As reliance on AI grows, defending critical AI infrastructure incl. nuclear power plants, renewables, energy grids and huge data centres from cyber threats will become paramount. Robust cybersecurity measures are essential to protect the systems that power AI innovation.
Stay True to Your Values Amid rapid growth and external pressures, businesses might loose their original values. I personally think, they should never do that. Of course, they have to adapt, but they also must remain loyal to their founders, employees, and core values. Selling out for short-term gains can undermine long-term trust and stability. Sam Altman's move might be led by his long-term vision to ensure AI supremacy for the U.S. resulting in worldwide democracy and abundance, but giving such a powerful tool and government support for-profit organization, could dramatically reshape the power balance on the entire planet.
Conclusion
OpenAI's strategic evolution reflects a comprehensive approach to navigating the complexities of the AI landscape. By focusing on innovation, collaboration, and ethical considerations, OpenAI is still positioning itself as a leader capable of competing on a global scale. As articulated in their latest press release,
"Our mission is to ensure that artificial general intelligence benefits all of humanity."
So be it.
Ex-Google CEO Eric Schmidt: "Pull the Plug!"
In a recent discussion on one of my favourite podcasts on Youtube, The Diary of a CEO by Steven Bartlett, Eric Schmidt, the former CEO of Google, shared some thought-provoking insights on artificial intelligence and its implications for society.
One of my favorite moments was when he emphasized the importance of knowing when to "Pull the Plug." He outlined specific scenarios where this might be necessary: when AI systems start to communicate in a language we don't understand, when they start developing themselves autonomously, or when they show activities that can lead to significant societal harm (e.g. designing zero-day-attacks, developing viruses).
On the other hand, Schmidt is quite optimistic about AI. Yes, he wants more regulation and governmental actions, but in the long term he believes, that AI will help humanity to overcome current global challenges.
Schmidt also highlighted his book, Genesis, co-authored with the late Henry Kissinger, where they explore the profound impact of AI on our future. I can hardly wait to read the book myself, since it is surely offers a sophisticated, but overall optimistic approach to AI and the future of humanity.
You can pre-order the book now, it will be available on 19. November 2024. 🚀🤖✨
NotebookLM
NotebookLM was 'the ChatGPT moment of 2024' for me.
Just listen to The Augmented Manager Podcast #5, where the two hosts discuss the main article of this very Newsletter, and you'll understand.
I simply copied the text of the Battle for AI Supremacy into NotebookLM, and it generated this lively conversation from it within 1-2 minutes.
NotebookLM's Audio Overview feature transforms your documents into engaging audio discussions between two AI hosts, providing a conversational summary of your material.
Here is how you can try it:
Access NotebookLM: Navigate to NotebookLM and sign in with your Google account.
Create a New Notebook
Select Source: You can upload any text, video, website, sound in multiple formats.
Generate an Audio Overview: Customise it using text prompts
Listen and Share: The sound file will be available for download or share in 1-2 minutes (currently English only).
For a visual guide, check out Google's official site: Google Blog
AI-generated Digital Twin of the St. Peter’s Basilica
The Vatican, in collaboration with Microsoft, has unveiled an innovative AI-generated digital twin of St. Peter's Basilica, aimed at enhancing visitor experiences and supporting conservation efforts. This project was officially launched on November 11, 2024, in anticipation of the Vatican's 2025 Jubilee, which is expected to attract over 30 million pilgrims.
Visitors look at AI-generated images of corridors and rooms inside St. Peter's Basilica as the Vatican presents a new AI-enhanced experience developed in collaboration with Microsoft (Source: Reuters)
Key Features of the Digital Twin
Immersive Virtual Tours: The digital twin allows for virtual visits to the Basilica, providing an interactive experience that includes access to areas typically off-limits to the public, such as the Roman tombs beneath the structure and intricate artwork in the high dome
Advanced Technology: Utilizing over 400,000 high-resolution images captured through drones and laser scanning during a month when the Basilica was closed, the project encompasses 22 petabytes of data, equivalent to five million DVDs
Visitor Management: A novel feature of this initiative is the ability for visitors to reserve entry times, addressing long wait times typical at one of the world's most visited sites
Pope Francis expressed his support for the project, advocating for ethical AI that aligns with human values such as compassion and morality. He noted that technology should serve to both preserve cultural heritage and facilitate access to spiritual experiences.
Prince in the Purple Rain
I brought you two AI generated images today. The first was generated by DALL-E using a simple prompt: "Prince in the Purple Rain with his iconic white guitar, watercolor painting."
I only had to do minor adjustments in Photoshop, and the image was ready.
Prince in the Purple Rain by DALL-E
The second one was made in Canva, using the original image and Canva's "magic expand" feature. The built-in generative AI filled the empty areas in no time
.Prince in the Purple Rain Expanded with Canva
A simple prompt led to a pretty cool result. So cool, in fact, that this single digital image (the first one) became the most popular item in my hobby side hustle Etsy shop. With 48 sales, it has generated 448 EUR in revenue over the past few months.
As an AI and tech enthusiast, I set up the shop to test the heavily marketed idea that you can generate passive income by selling AI-generated digital products. I won't get rich from it for sure, but I can pay for my ChatGPT Plus, Midjourney and Canva licenses, and play further... Crazy, right?
Midjourney Image Generation
Midjourney generates images based on text prompts. To create an image, you enter the /imagine command followed by your descriptive prompt and parameters.
Midjourney offers various parameters—denoted by double dashes (--)—that you can append to your prompts to customize the generated images. Here's an overview of some commonly used parameters:
Aspect Ratio (--ar or --aspect): Adjusts the width-to-height ratio of the image. For example, --ar 16:9produces a widescreen image, while --ar 1:1 yields a square image.
Stylize (--s or --stylize): Controls how strongly Midjourney applies its artistic style to the image. Higher values (up to 1000) result in more stylized images, whereas lower values adhere more closely to the prompt details. The default value is 100.
Version (--v or --version): Choose the Midjourney version for image generation.
Chaos (--chaos or --c): Influences the variability of the generated images. Higher values (up to 100) produce more unexpected and varied results, while lower values yield more consistent outputs.
and more.
I applied the same prompt that I used for DALL-E, Midjourney added its default parameters to it.
Prompt: /imagine Prince in the Purple Rain with his iconic white guitar, watercolor painting --s 750 --v 5.0
Result:
Opinion: Midjourney always gives you 4 results at first, you can select which one to enhance further and upscale. Midjourney typically gives much more detailed results, than DALL-E, but sometimes less is more.
What do you think?
“When deploying AI, whether you focus on top-line growth or bottom-line profitability, start with the customer and work backward.”— Rob Garf, vice president and general manager, Salesforce Retail
That's it for this weekend. Thanks for reading!
✌️Until next time!
The Augmented Manager
🎙️Listen to The Augmented Manager Podcast on: Substack, Spotify and Apple Podcasts.
🔄 Check out all previous newsletters, deep dives and resources here.
If you like The Augmented Manager, spread the word - invite friends to follow!