KW 39: Nvidia wants to buy ARM, Foundation uses AI to root out antisemitism, Federal AI Association not satisfied with von der Leyen’s speech


Nvidia wants to buy ARM: Nvidia is buying British chipmaker ARM from SoftBank for as much as $40 billion — a total that would make it the largest semiconductor deal ever. The move would allow Nvidia to become an industry leader across more connected devices, including smartphones, PCs, robotics and 5G, CEO Jensen Huang wrote in a letter to employees. Huang cited Arm’s experience in finding ways to make hardware that can run sophisticated A.I. programs while consuming far less energy as a key asset Nvidia was hoping to exploit across its portfolio of AI-dedicated chips in the future. He also said Arm would allow Nvidia to reach a whole new set of customers beyond its traditional datacenter and gaming markets.,

Foundation uses AI to root out antisemitism: The Alfred Landecker Foundation, which was created in 2019, has invested three million euros to fund the development of an antisemitism AI. Researchers, computational linguists and historians are working on the project to identify certain codes for antisemitic messages in German, English and French. The aim is to develop an instrument for website operators that enables them to easily and accurately identify hate speech and agitation.

AI isn’t threatening human laborers yet: While there are ever greater advances in the field of artificial intelligence and there are fears that humans will be gradually replaced by machines, technology is currently still far from that point. One example of this is GPT-3, an autoregressive language model that uses deep learning to produce human-like text. It is touted as a high quality, modern text machine. In fact, it manages to elegantly formulate a number of meaningful and correct sentences – but there are still major weaknesses in terms of content. The AI imitates language, it doesn’t think about the content and doesn’t formulate its own assumptions.

– Advertisement –
Blockchain – Ticker – With our weekly newsletter we provide you with the most important developments in blockchain technology and scene.

Federal AI Association not satisfied with von der Leyen’s speech: Last Wednesday, EU Commission President Ursula von der Leyer spoke about the state of the EU. She made it clear that AI had to be regulated. But the German government’s Federal AI Association believes that the Commission president’s plans are falling short. The association, for example, calls for a gradual implementation of regulation: “Any AI application that can cause physical harm to humans or animals should be exposed to significantly higher regulatory hurdles than AI applications that do not,” it said in a statement. Proportionality must be maintained so that there is no damage to the technology’s development.

Chatbots don’t have promised quality: In the opinion of AI expert Peter Gentsch, the requirements that are placed on AIs are often too high – this has a lot to do with wishful thinking. In marketing, experts hope for chatbots that can permanently supply customers – in practice, however, most AI systems are still weak. Gentsch expects that the level of AI that would make chatbots really intelligent will only be reached in about ten years. Customers are willing to use chatbots – but they want to have a competent system.

US-China: “Those who lead in AI dominate economically and militarily”
Research: Artificial intelligence needs people
Medium-sized companies: Framework conditions slow down AI
Production: How artificial intelligence optimizes quality control
B2C: Don’t be afraid of artificial intelligence in customer service

– Advertisement –
IoT – Ticker -The physical world meets the digital one. Internet of Things as an interface that revolutionizes both the industry and everyday life. Get a weekly update from the world of “Internet of Things”.


In Germany, 247 startups are currently making significant use of artificial intelligence.


Money laundering association calls for AI support: The German ministry of justice and consumer protection and the finance ministry want to fight money laundering more intensively in Germany and have already submitted a first draft law. “It is important to strengthen supervision and enforcement and to consistently develop the application and implementation provisions on an empirical basis,” said Christian Tsambikakis, Chairman of the Federal Association of Money Laundering Officers. The data sets and information that would be collected as part of the anti-money laundering campaigns can hardly be controlled without AI, as some of these are transnational networks that are targeted by investigators.

AI reaches its limits with Covid: AI researcher Gary Marcus sees the Covid-19 pandemic as a wake-up call for AI. Despite a long history of research and development, it has become clear that AI is reaching its limits and is not achieving the hoped-for effects in combating the pandemic. Marcus demands that the technology no longer be used primarily for marketing and advertising, but for areas where it can make a real difference. In his book “Rebooting AI”, he calls for a return to rule-based research.


AI to combat child pornography: A system is being developed in Germany that filters out internet content based on pornographic patterns. The plan is to use this new AI system in Lower Saxony in the State Criminal Police Office to combat child abuse and child pornography. The aim is to identify illegal activities against children at an early stage. Several other federal states have already expressed an interest in the system in order to use it themselves.


“Machine learning will increasingly be used to analyze how users deal with different tasks. Once patterns are identified, they can be automated – with the technologies learning and improving themselves. For example, an AI can recognize that an employee has carried out an action 20 times in the same way and ask whether the action should be automated from now on. This reduces the number of routine tasks that are often perceived as boring, and employees have more time for qualified tasks.”
Maximilian Thost, Country Manager DACH at Quinyx, on the use of AI in human resources.


Big Brother Award given to companies with problematic data handling: The questionable honor of Germany’s Big Brother Award is awarded to companies and organizations that develop applications that are controversial under data protection law. This year, Tesla is also included on the list of award recipients – the company has installed a number of outdoor cameras in its cars that scan and analyze the environment. This data is sent to servers in California. The award jury commented on Tesla’s technology as follows: “We have nothing against vehicle assistance systems, and nothing against semi-automated driving. This requires sensors and so-called artificial intelligence. But from a data protection point of view, this data can and must largely remain in the car.”,

Newsletter subscription

Subscribe to our free weekly newsletter for a compact overview of Artificial Intelligence topics:


More digital news briefings

Our political briefings