Techstrong News Recap – May 11, 2023
On May 11, William Willis covered China’s first arrest over fake news created by AI, Anthropic releasing principles to govern its “constitutional AI” and an EU draft that would change cybersecurity labeling rules for U.S. big tech companies. He also talked about LinkedIn cutting 700 jobs and shuttering InCareer. Finally, he previewed articles from Security Boulevard, Digital CxO and DevOps.com, which covered a new Magecart skimmer scam, health care 5.0 and AWS releasing its Cedar policy language tool as an open source project.`
Transcript
Hi again, everyone here the headlines from a11. First up Chinese authorities have made one of the first arrests under the new anti-ai guidelines to prohibit the misuse of artificial intelligence services to distribute false information. The suspect surnamed Hong was detained in the Gansu Province for allegedly using chat GPT to generate fake news articles.
Hung is accused of rewriting existing viral news articles and posting them online to make money through internet traffic in the fake articles were viewed over 15,000 times before being removed. Hung was charged with the pinking picking quarrels and provoking trouble a catch-all offense that is often used to arrest individuals accused of creating and spreading misinformation. Critics of widely criticized the offense for its potential to limit free speech and arrest government critics zooming out while many countries are apprehensive about AI China is one of the few countries that has completely blocked access to chat GPT and this arrest shows China's determination to put a leash on AI technology.
Next anthropic an AI startup founded by former openai employees recently revealed its written principles towards AI to follow which draws from various sources including the un's universal Declaration of Human Rights and apples terms of service. The company has been working on a way to make AI safe through constitutional AI a method of training AI systems to allow the AI system to manage itself and reduce the Reliance on human moderators to rate its output for things like hate speech and toxicity. The document includes guidelines to encourage Freedom equality and prevent racism sexism and discrimination and considered on Western perspectives.
With all us being said though the AI Industries already grappling with perceived bias and chat pots. But this constitution sets the minimum minimize that as well as toxicity on anthropics own chatbot and pave the way for other AIS to follow. In other news a new EU draft states that Amazon Google Microsoft and other non-european Union cloud service providers seeking an EU saver security label.
To handle sensitive data must enter a joint venture with an eu-based company additionally US tech Giants can only have a minority stake in the Venture an employees with access to EU data must undergo specific screening and be located in the 27 country block. The cloud service must also be operated and maintained from the EU in all customer data stored in the region. The EU certification scheme would vouch for the cypressecurity of cloud services and determine how governments and firms in the block to set a vendor for business as well as also working towards the eu's goal of continuing to rein in us bigtac next up LinkedIn the professional networking platform announced that it is laying off of her 700 workers and closing its China jobs app in career marketing the most recent set of tech company layoffs the company cited shifts in customer behavior and slower Revenue growth as the reasons for the cut.
This being said though, they say that number of job losses could be less than 500 as part of a strategic shake-up LinkedIn is set that will be opening over 250 roles as part of its operations Team new business and accounts management teams. The local jobs app for China in Korea. However, will be shut down by August citing Fierce competition and the challenging macroeconomic climate.
On Security Boulevard, we have an article that looks at a new Mage cart skimmer scam that dupes the stores check out page. The scam uses a customized web element to make it harder to distinguish a fake form from a real checkout page according to Malwarebytes. Well using a modal is not new.
com on digital cxl. 0 or digital Wellness allows for new opportunities to improve health services worldwide. This being said though.
0? com. com.
We've an article looking at aws's announcement that it is making it Cedar policy as a code tool and open source project. com. And that's today's text wrong news recap.





