AI botches news articles

Welcome back to The Daily Update — OpenAI is under fire again as a detailed complaint was filed with the Polish data protection authority. We’ve also got an epic AI newswriting fail on deck. Here’s the full menu:

🗞️ The Latest in AI: OpenAI accused of data protection breaches

🤖 AI Tool of the Day: Research confidently with Perplexity

📸 Business Spotlight: Gannett to pause AI experiment after botched articles

🚨 AI Roundup: Four quick hits

Read time: 2 minutes

🗞️ The Latest in AI: OpenAI accused of data protection breaches

Privacy researchers are raising questions about OpenAI’s ability to comply with EU privacy rules. A detailed complaint filed with the Polish data protection authority alleges that the company is in breach of the bloc’s General Data Protection Regulation (GDPR).

Key points:

• Lukasz Olejnik, a security and privacy researcher, filed the 17-page complaint after using ChatGPT to generate a biography of himself that contained some errors.

• Olejnik contacted OpenAI to ask for the inaccurate information to be corrected and to provide information about ChatGPT’s data collection.

• The complaint argues that OpenAI failed to produce all the information it must by law, including information about its processing of personal data for AI model training.

• OpenAI is accused of breaching several GDPR provisions including lawful basis, transparency, fairness, data access rights and privacy by design.

Why it matters: This news comes a day after The Information reported that OpenAI is on pace to generate more than $1 billion in revenue over the next 12 months. Confirmed violations of the GDPR can attract fines as high as 4% of global annual turnover.

More details on the complaint against OpenAI here.

🤖 AI Tool of the Day: Research confidently with Perplexity

Perplexity AI is a search interface that combines GPT-3.5 and Microsoft Bing to answer any question you ask.

Most people use chatbots to quickly learn about new topics, but AI chatbots are known to hallucinate and generate fabricated information. Perplexity sets itself apart by linking sources to the information it provides.

You can also personalize search results by creating an AI Profile that includes your location and some information about yourself.

If you’ve been using chatbots as research and search tools, consider making the switch to Perplexity. Try it for free here.

📸 Business Spotlight: Gannett to pause AI experiment after botched articles

Newspaper chain Gannett has paused the use of an AI tool to write high school sports articles after the technology made several errors in at least one of its papers.

Key points:

• Several high school sports reports written by LedeAI and published by the Columbus Dispatch went viral on social media this week.

• Many reports featured identical language describing “high school football action.” Articles frequently noted that one team “took victory away from” another and described “cruise-control” wins.

• The reports were mocked for generally sounding like they were written by a computer with no knowledge of sports.

• Gannett laid off 6% of its news division in December.

The bottom line: It’ll be a while before AI can write this newsletter. Gannett was moving too hastily here.

🚨 AI Roundup: Four quick hits

• Play Safe: Call of Duty will use AI to moderate voice chats.

• ‘Empowering Associates’: Walmart will roll out a generative AI workplace app to roughly 50,000 employees.

• Ready, Set, Go: An AI quadcopter has beaten human champions at drone racing.

• Smart Meals: Samsung launches a meal planning and recipe discovery platform called Samsung Food. The platform includes personalization features through “Food AI.”

Have questions or thoughts on today’s newsletter? Reply to this email and I’ll get back to you as soon as I can. Thanks for reading The Daily Update!

Jack

(P.S. If you know someone that may find today’s newsletter useful, send it their way!)