- Elon Musk-owned AI firm xAI has released the latest versions of Grok AI – Grok AI-2 and Grok AI-2.
- Both are available in beta mode as of now and by the end of this month will be available through the AI developer’s enterprise API.
- The most interesting feature is the newly added image generator which has already raised concerns owing to its lack of rules for creation.
Elon Musk’s AI firm xAI has launched two new models of the GrokAI namely Grok-2 and Grok-2 Mini. It’s currently only available in beta mode. And by the end of this month, both models will be available through the AI developer’s enterprise API.
The two new models offer better performance, improved reasoning, and most importantly, it can now generate images too which can be directly posted on X.
‘We are excited to release an early preview of Grok-2, a significant step forward from our previous model Grok-1.5, featuring frontier capabilities in chat, coding, and reasoning.’ – xAI
It also talked about the difference between the two new versions. Grok-2 Mini is apparently a smaller yet equally capable version of Grok-2 and is best known for wonderfully balancing answer quality with speed.
Last but not least, both versions are being trained to improve search capabilities, reply functions, and post analytics.
The company also said that it will soon release a preview of multimodal understanding as a core part of Grok’s experience on X and API.
This could mean that Musk is planning to integrate them with X to introduce AI-powered replies soon.
Grok isn’t like the usual chatbots that follow a question-answer pattern. Instead, Grok’s responses are witty and sarcastic. Musk has also gone on to say that Grok isn’t for everyone – only people with a sense of humor can use Grok well.
How Well Do the Image Generators Work?
Image generation with AI has always been controversial. Especially because almost all AI platforms before this (take Google’s Gemini for example) have landed in quite a few controversies over inaccuracies or potential risks. Grok’s latest versions are no different.
In some of the examples that surfaced online, it’s clear that there aren’t many rules surrounding what kind of images a user can or cannot generate.
Spreading misinformation would become easier than ever. Users with malicious intent might also be able to create hyper-realistic deep fakes of their victims and blackmail them.
- We have already seen how quickly the sexually explicit deepfake images of Taylor Swift spread on the internet.
- And only recently, Apple had to remove 3 apps from the App Store after they were found generating sexually explicit images without consent.
If GrokAI isn’t capped with usage rules. it would only lead to more misuse of AI in the coming few years.
Our Editorial Process
Our Editorial Process
Share 
Krishi Chowdhary Journalist 
Question & Answers (0)