What are the Opportunities, Risks, Challenges and Potentials of Artificial Intelligence (AI)?
The hype about Artificial Intelligence (AI) is not new. In the 1950s and 1970s, first steps in artificial intelligence such as the chess-playing computer fired the imaginations on what might be possible in the very near future. But why is the AI trend exactly now so hot again? This is essentially due to the fact that computer power and storage capacity have risen enormously. Moor‘s Law already implied that computer power would double every 18 months. And exactly this computer power is the necessary requirement for collecting, processing and analyzing more data points. The significant increase in the number of processed data points creates a higher precision of evaluations, so that new patterns and laws can be identified. Therefore, the significantly better processing and quality of the data enables better analysis results and completely new applications. As a result, artificial intelligence now has the potential to become a real game changer and to reshape the future fundamentally. It is only a matter of time before artificial intelligence will become an integral part of daily work processes.
Actually, artificial intelligence is already an integral part of many workflows today. This ranges from search engines and purchase recommendations to assistance systems. “Predictions” are probably the best known application of artificial intelligence, along with automated recognition of image patterns or building artificial scenarios. In the future, however, artificial intelligence will mean much more than automated conclusions that can be drawn from large amounts of data. In the future, algorithms will increasingly optimize themselves. But it is still a long way until fully automated robots will dominate the world. In many industries, the focus will not be on fully autonomously controlled robots, but rather on the interaction between man and machine.
Changes will first become visible in industries whose standardized processes and work tasks can be described easily with alogrithms. These industries will notice a clear benefit for their performance when evaluating large amounts of data and enabling improved analysis. These include industries such as banking and automotive, but also medical diagnostics. Medical assistance services as well as applications for work safety in production are good examples here. Human errors or false diagnosis in medicine could also be reduced with the knowledge of big data. In addition, autonomously controlled vehicles could significantly increase safety compared to human errors when driving a car. So in many areas, algorithms can surpass human experience through data accuracy. But in areas that require high social competence, very flexible movements or creativity, on the other hand, artificial intelligence cannot fully map the complexity of human behavior yet.
In the public discourse, AI usually focuses on the impact AI has on employment and is therefore often critizied. Here it is important to differentiate the effects artificial intelligences really has. The influence of technology on production and employment is not new. And also the influence of the interaction between technology and people has also always existed. One can distinguish between technologies that increase productivity and technologies that automate processes and thus replace human work force. Artificial intelligence, which makes it possible to carry out work that was previously impossible, can increase turnover and wage levels. Whereas applications of artificial intelligence, which are more likely to create efficiencies and thus replace human labor, always have a cost-cutting effect and therefore a direct impact on the employment. Though in reality, the effect of technologies based on artificial intelligence will not only have a single effect. In the best scenario, increasing productivity and cost savings are interlinked. But weak technologies create only cost savings via efficencies, but have no effect on productivity. How artificial intelligence actually affects employment, however, depends on many more factors. For example, the growth potential of the specific markets or industries. So it makes sense to look at the opportunities artificial intelligence has or has not in every single segment individually where it could be applied.
It is fundamental to understand that artificial intelligence is not about replicating a person’s thoughts and actions, but about reducing the complexity of specific tasks. However, the development of artificial intelligence still poses many challenges. These include both the influence that the developers themselves have on artificial intelligence as well as how the training data is interpreted. Prominent examples are recruiting algorithms that were based exclusively on data of white, male role models and thus only made a discriminating selection. However, not only the structure and the data basis of the algorithm, but also its use must be considered. In China, for example, cameras capture a large share of the public space. These recordings create a high data volume and give them consequently a high accuracy of algorithms interpreting individual facial features. This leads to the remaining question if AI can be misused for politcial or other interests. But this also demonstrates how important it is to gain a lead position in artificial intelligence.
A guest contribution by Dr. Tanja Emmerling, Principal at High-Tech Gründerfonds and Head of the Berlin office. Since 2014, she has managed her own portfolio in the software sector and focuses on the opportunities of artificial intelligence, big data and IoT. Before joining High-Tech Gründerfonds, she spent more than six years in the media industry, most recently as Head of New Ventures building incubation projects. Dr. Tanja Emmerling studied economics and languages and accompanied innovators upon entering new markets early on in her career. Additionall, she has participated in executive programs for new business and lean start-up at Harvard and Stanford.
To find out more about Porsche and Technology, follow us on Twitter, LinkedIn and Instagram.
