Here is the top trending news from the world of technology. News that every tech enthusiast should keep a tab on.
1)
Apple kills Epic Games Developer Account, calls it a ‘threat’
The battle between Apple and Epic Games took an interesting turn on Wednesday, with Apple terminating Epic Games’ developer account. This effectively prevents Epic Games from publishing any apps or games on the iOS App Store. Apple has justified its action by pointing out at what it claims is Epic’s ‘untrustworthy behavior. It further claimed that Epic Games’ repeated violation of App Store guidelines posed a serious “threat” to its iOS ecosystem. This latest development is likely to further escalate the legal battle between Apple and Epic Games. It also once again raises broader questions about the power and control that large tech companies like Apple wield over app ecosystems
2)
Copilot Designer is risky & harmful tool, warns Microsoft Engineer
Shane Jones, a Microsoft engineer who worked with Microsoft, recently wrote a letter to FTC raising concerns about company’s AI tool Copilot Designer. Jones, who was part of the Copilot Designer’s testing team, has claimed that the AI tool is prone to generating violent and sexually explicit images. He further added that the tool generated images that blatantly violated the copyright laws. Jones brough these concerns to Microsoft’s notice. While the company acknowledged his concerns, it refused to withdraw Copilot Designer from the market. This incident has sparked discussions about the need for stricter safeguards and regulations for AI tools.
3)
Why is Meta planning to build a giant AI Model
Facebook and Instagram’s parent company Meta is planning to build yet another AI model. However, this AI model would be the largest that the Meta has ever built. It will be powering and managing its entire video ecosystem across all platforms, encompassing both short-form videos like Reels and longer videos. The company is hopeful that this will lead to a more engaging and relevant user experience for users. Currently, the information available about this giant AI model is scarce but Meta is reportedly planning to launch this AI model by 2026. Over all, Meta has been spending billions of dollars on AI and this latest project again signifies Meta’s commitment to build cutting edge AI technology.
4)
OpenAI’s GPT4 performed worst in copyright infringement test
Patronus AI, an AI evaluation company founded by ex-Meta researchers, recently tested four popular AI models for copyright infringement. These four popular AI models included OpenAI’s GPT-4, Anthropic’s Claude 2, Meta’s Llama 2 and Mistral AI’s Mixtral. Among these models, GPT-4 generated the highest amount of copyrighted content, responding to an average 44% of prompts requesting for text from books with the copyrighted text. This meant that GPT-4 performed worst among all the AI models. OpenAI has so far not commented on this test’s results. Overall, copyright infringement has been among the most hotly debated topics in the AI industry. All big AI and tech companies have been accused of plagiarizing and copying the content of writers and publishing companies without their consent.
5)
Ex Google engineer allegedly stole AI secrets for a Chinese company
A former Google engineer called Linwei Ding was arrested earlier this week for allegedly stealing AI trade secrets from Google. Ding, who is a Chinese national and started working with Google in 2019, allegedly wanted to share this trade secret with a top Chinese tech company. However, the name of this Chinese company has not been identified. According to preliminary information, the Chinese national had uploaded trade secrets from his Google-issued laptop to his personal cloud storage accounts. This incident had come to light at a time when America and China are involved in a AI war. Only last month, the Biden administration had banned the export of advanced chips to top Chinese companies.