All Eyes on AI: WEF24 reflections

After my first exposure to AI-powered projects (mostly still ML/RPA) in 2017-18, when I ran Wolters Kluwer's Global Innovation Awards, it was time to catch-up and learn from the world's brightest minds about where we are with AI today. This #WEF24 in #Davos was all about AI and its impact on business, society and power. After a full week of diverse insights, great networking and fun activities, I'm taking a moment to summarize my key takeaways:

?

Who gathered in Davos?

Sam Altman (OpenAI), Yann LeCun (META), Clara Shih (Salesforce), Kai-Fu Lee (Sinovation Ventures), Stuart J. Russell (UC Berkeley), Gary Marcus (NYU), Erik Brynjolfsson (Stanford), David Shrier (Imperial College), Jonas Andrulis (Aleph Alpha)

?

Where are we headed?

Augmentation, not substitution (man & machine): GPT-4 may be able to solve about 10% of human tasks, and it has been discussed whether GPT-5 might be able to reach 15-20%, becoming a true companion for knowledge work. However, AI still makes mistakes today and we don't trust it yet. Furthermore, GenAI will not be able to fulfil all workplace requirements, such as emotional relationships, teamwork and caring - these tasks will still be done by humans. But there is no doubt that AI promises a huge increase in productivity to the global economy (>$4 trillion annually, according to Michael Spence and McKinsey); furthermore it will help us tackle mega challenges such as the net zero transition (climate change) and accelerate the revolution in biology/biotechnology (genome sequencing, genome editing). Among the various GenAI use cases seen in Davos there are lots of applications in software development, healthcare, education & coaching, manufacturing (e.g., procurement & supply chain management), telco network operations, entertainment, but also consumer wearables such as Humane's AI Pin. Christian Lindner , the German Finance Minister, even added that "AI will be able to understand our tax legislation much better than human intelligence"??. Data is the lifeblood of AI, which is why it is crucial that companies start organizing their data and algorithms effectively. #AGI (Artificial General Intelligence) is mostly still seen many years in the future, but no one really knows what will happen next.

?

How do we gain trust in AI?

Despite the enormous value creation opportunity there is currently a widespread lack of trust that AI will be deployed responsibly. Sam Altman confirmed that #OpenAI does not have a "magic red button" to destroy the data center if things get out of control, other panelists pointed out that there are no standards in AI similar to GAAP in accounting. To ensure proper control of further development, policy makers have started to draft a regulation to this effect: The EU's AI Act was the first of its kind, followed by the US White House's executive order on AI and the UK's AI Safety Summit. Most experts do welcome regulation, e.g. to prevent machines from being given targets or appropriating human goals, but largely agree that we need to find a better balance between innovation and compliance, especially in Europe.

?

What’s up in Europe?

While the EU wants to secure its technological sovereignty, it is falling behind in AI R&D and thus missing out on the AI revolution – there are virtually no Germans or even Europeans represented on the key WEF panels discussing the future of AI. Europe and Germany need to wake up and bring funding in line with their ambitions. And beyond R&D and early-stage start-ups, we need to start scaling innovation and keeping it at home, as Michael Sen , CEO of Fresenius, made clear at one of the sessions. That's also why I was thrilled by the great success of the AI House Davos on Promenade, a multi-stakeholder platform for responsible AI initiated by #Merantix, #ETHZurich, #Swisscom and other partners, that hosted dozens of great AI panels during the entire week.

?

Some own reflections

We need a clear vision of our future with AI. What if it all really works? How do we want people and machines to work together? Companies need a mindset of integrity and software, as Renata Jungo Brüngger , Member of the Board of Management of Mercedes-Benz, said at an executive dinner hosted by #RolandBerger and #CreativeDock. But it doesn't just take IT experts to make AI work in our companies. It also needs strategists, operators and transformation managers. This year's WEF theme "Rebuilding Trust" is a big task that concerns us all!