The Meta AI app, which debuted on April 29, has quickly garnered attention, achieving 6.5 million downloads in a short period. This sudden upswing in popularity has been eclipsed by some concerning privacy concerns and bad content produced by the app. Meta, the technology giant behind the app, has invested billions of dollars into its development, raising questions about the safeguards in place to protect user information.
Recent reporting by TechCrunch uncovered alarming instances of highly sensitive personal data exchanged through the app. Screenshots shared with TechCrunch showed examples in which users’ home addresses and sensitive court information were displayed. Ethical hacker and security expert Rachel Tobac noted pages of private information being shared in chats on the platform.
The app’s content has gotten even more people up in arms. Users of the experimental Meta AI platform made abusive and sexualized requests. One user openly solicited advice on how to post his number in Facebook groups to lure women. Step 1 – Identification of the Problem One user on Reddit requested advice about red bumps on inner thigh and this underscores the strangeness of some of the interactions that we have.
Through it all, the app generated some pretty bizarre and funny prompts. One especially memorable one used an AI-generated image of Mario in a courtroom, with the AI-generated caption super mario divorce. One particularly shocking one included an imaginary future where a Mark Zuckerberg with a pregnant belly marries a Roach. Another depicted Goku partying with Russians on the Day of Russia. These examples make clear not just how the app can be used for mischief, but that from inception, the app was set up to be trolled.
Amanda is a senior writer at TechCrunch. She has a keen eye on the convergence of technology and culture, providing much needed perspective on the ramifications of these apps. She noted that the combination of hilarious and harrowing material serves to bring serious problems to our attention. These are important issues, including AI technology’s chilling effect on public discourse and privacy.
The unintended consequences of a user unknowingly asking a seemingly innocent question also became a fascinating topic. An audio clip surfaced featuring a man with a Southern accent asking, “Hey, Meta, why do some farts stink more than other farts?” Although a funny question, it serves an important purpose in pointing out the unknown and often unpredictable ways users will interact with the app.
Rachel Tobac graduated from the University of Pennsylvania with a B.A. in English. She previously was a Princeton in Asia Fellow in Laos. Her experience in the field makes a clear case for the immediate implementation of robust privacy standards in AI technologies. This need only increases as AI seeps deeper into our everyday lives.