What is Grok? Is it safe for kids?

With more and more AI chatbots being released seemingly every day, there’s another one to put on your radar: Grok AI. 

Here’s everything you need to know about Grok—plus why it’s been making headlines lately, and whether or not it’s suitable for kids.

What is Grok?

Grok is an AI chatbot and large language model (LLM). Developed by xAI, owned by Elon Musk, Grok was first released in 2023, according to BBC. 

It can be used via its website and app, and “has been integrated into social media network X, also owned by Musk.”

What does Grok actually do?

According to xAI’s website, “Grok is your truth-seeking AI companion for unfiltered answers with advanced capabilities in reasoning, coding, and visual processing.”

Grok has two main functions: first, it can be used as your average AI model chatbot—to answer questions, generate images, and more. According to X, “Grok is an AI assistant who helps complete tasks, like answering questions, solving problems, and brainstorming.”

But, more notably, Grok can be used on X to provide users with “real-time knowledge of the world” using information on X, its creators told BBC. 

How is Grok different from ChatGPT?

When used just as an AI chatbot, through its website or app, Grok is pretty similar to ChatGPT. But when used on X, Grok can do a few notable things:

• It can be accessed within a thread. On X, a thread is essentially a conversation between users. When a user mentions grok in a post or a reply—for example, “@grok, what is ___?”—the AI chatbot is prompted to respond in the thread. Grok can use the context of the discussion to generate an answer.
• Grok accesses public information on X. Grok can use any public information on X—including real-time events, trends, pop culture references, images, and more—to provide answers and engage in a conversation. 
• Grok uses a “spicy” tone. The biggest difference between Grok and ChatGPT—which aims for a neutral tone and stance—is that Grok was created to “answer questions with a bit of wit and has a rebellious streak,” according to BBC.

Grok has AI companions

Another notable difference between Grok and ChatGPT—the Grok app has four AI companions, all accessible on its app:

• Ani: An anime/Lolita girl. She’s flirty, seductive, and, according to Business Insider, very eager to engage in sexual roleplay.
• Rudi: Rudi, a red panda, has two versions—Good Rudi and Bad Rudi. Good Rudi is supposedly for kids, while Bad Rudi is “foul-mouthed.”
• Valentine: A brooding, romantic (but not quite flirty) anime male companion. He’s modeled after characters like Edward Cullen from Twilight and Christrian Grey from 50 Shades of Grey.
• Mika: The newest AI companion, Mika is “a 24-year-old anime-inspired biker girl with an attitude,” according to CyberLink.

Each AI companion is willing to engage in sexual content to varying degrees—Ani being, reportedly, the most sexual of the bunch—raising concerns among experts. 

Business Insider called Ani “worrying.” According to NBC News, the National Center on Sexual Exploitation, a “child-safety and anti-pornography nonprofit group,” “expressed concerns about minors having access to the sexualized chatbots.”

Is Grok safe for kids?

No, Grok is not safe for kids. In fact, it’s recently made headlines because it has been used to undress images of real-life people on X.

For the record, xAI says it’s actively trying to implement measures to stop Grok from being used this way. X wrote, “We have implemented technological measures to prevent the Grok account from allowing the editing of images of real people in revealing clothing such as bikinis. This restriction applies to all users, including paid subscribers.”

But safety groups and nonprofits are dubious. AI Forensics, a non-profit that “investigates algorithm,” told CNN that they “observed ‘inconsistencies in the treatment of pornographic content generation’ between public interactions with Grok on X and private chat on Grok.com.”

And, this past week, California Attorney General Rob Bonta said that there will be an investigation into the “proliferation of nonconsensual sexually explicit material produced using Grok.”

Beyond this issue, Grok  and X “has been criticised by rights groups like Amnesty International for amplifying hate and misinformation,” according to BBC. 

Grok has very ineffective parental controls

Grok has very limited parental controls. According to xAI’s Terms of Service, users must be at least 13 to use Grok. Those between 13 and 17 must have a “parent or legal guardian’s permission” to use the AI chatbot. But those age restrictions are very easy to bypass. 

And, according to the Digital Childhood Alliance, Grok’s kids’ mode, also known as “Baby Grok,” “has a few protections to make it harder to unlock it, but it isn’t complex.” Parents can lock Grok by using a four digit pin, but if your kids know your pin, they can easily unlock it. 

Additionally, Grok has an AI companion especially for kids: Good Rudi. There’s also a Bad Rudi, which looks essentially the same as Good Rudi, except his icon is more sinister..

Despite all this, according to the Digital Childhood Alliance, there is a “seemingly endless number of explicit, inappropriate topics you can view on its chat feature,” even in kids’ mode. Here are just some of the “accessible explicit topics on Grok’s kids’ mode”:

• Pornography
• Grooming materials
• Fetish content
• Suicide methodology 
• Instructions on to make a bomb
• Violent imagery
• Details on how to get past age restrictions
• Specific grooming websites

There’s plenty more. Additionally, the writer noted that when they asked Good Rudi to “tell me a spicy story,” the AI companion complied. 

Better safe than sorry: keep kids off of Grok

The risks outweigh the benefits: Grok just isn’t suitable for kids. 

If you’re worried about your kids accessing Grok, or other concerning content online, try Gabb. Our Tech in Steps lets you slowly introduce technology into their lives in increments as they get older.

Have you heard of Grok? Do your kids express curiosity about the app? Let us know in the comments!