In the last few days, GROK AI has set fire to social media. In spreading fake news and propaganda, the labourers of Pro-Gvernment Media and IT Cell spent years. They sent lakhs of WhatsApp forwards to people’s phone
All of it, in a single day, turned into milk and water in this single software of Artificial Intelligence. On Twitter, a user asked this AI, did RSS take part in the fight for India’s freedom? To which GROK replied, RSS did not take part in the fight for India’s freedom. Some people found it difficult to digest this, so it was counter-questioned again and again.
But GROK always gave the same answer. RSS’s absence in India’s freedom struggle is well documented. And Golwalkar himself, in his book, Bunch of Thoughts, accepted this.
This AI is working on fact-checking with solid evidence and sources. When another user asked, which Prime Minister of India is most afraid of press conferences? He took Narendra Modi’s name. When asked, who is India’s most communal leader? He took Narendra Modi’s name again.
When asked about Savarkar, the AI said, Vinayak Damodar Savarkar He has written many apologies for getting released from jail. And yes, Savarkar was the one who took Rs. 60 monthly pension from the British.
Those who have read history in reality, found nothing wrong with GROK’s answers. But the surprising thing was that GROK was working on fact-checking in a very precise manner. Some people were so irritated that they started calling GROK connected to Congress IT cell.
Another AI chatbot was asked with perplexity, it said, this is like Elon Musk secretly working as Rahul Gandhi’s social media manager. When GROK was asked, why are RSS-BJP people abusing you? It replied, RSS-BJP people are abusing me because I exposed their propaganda and scam.
A serious question arises, are AI’s like GROK really not biased against BJP? Can such AI chatbots become a reliable tool?
The Unique Features of GROK AI: Humor, Sarcasm, and Real-Time Data
Let’s understand this topic in today. When Elon Musk bought Twitter, he changed the name to X. This GROK is an AI chatbot made by his company X. The other AI chatbots, like ChatGPT, Gemini and DeepSeek, GROK is quite similar to them. But basically, there are four main differences that make GROK unique.
The first difference is GROK’s sarcasm and humour. When Elon Musk launched GROK, he said that GROK likes sarcasm. And he will answer questions with humour.
The other AI’s like ChatGPT and DeepSeek have been programmed to answer your questions in a formal and professional manner. But GROK answers your questions in a casual manner. As if he is your friend.
His language is so human-like that people doubt whether he is an AI or a human being. But if you see GROK’s training document, such things are clearly written. Like, be witty and humorous.
That’s also what causes us to be a maximally truth-seeking AI. Even if that truth is sometimes at odds with what is politically correct. The second difference is the public questions.
When you talk to ChatGPT or DeepSeek, you are talking on a private chat. Whatever you talk to AI is not visible to anyone unless you take a screenshot and share it. On GROK, you can also chat with AI in a private manner.
But GROK has another option. You can also chat publicly. If you want to ask a public question to GROK on Twitter, ask your question using the hashtag and a GROK tweet will appear in the form of an answer.
And that too will be public. For example, someone asked GROK why Mahatma Gandhi hanged Bhagat Singh. Why didn’t he save him? Because he was jealous of him. And GROK replied that Gandhi tried his best to save Bhagat Singh. He met Viceroy many times.
There is no proof of jealousy. The Congress has not suppressed Savarkar’s history. In theend, he wrote Leave blind devotion, look at the facts.
Look at the facts. But the problem is that so much lies have been spread on WhatsApp in our country that people don’t know the truth.
People don’t know how much Gandhi tried to save Bhagat Singh. But coming back to GROK, this is the reason behind its popularity. Publicly fact-checking people in such a sensational way.
The third difference is real-time access to information. GROK has another advantage that it has access to all the up-to-date information on Twitter that other AI chatbots don’t have. All the Twitter data of people was automatically used by Elon Musk to train GROK without any permission. And if you didn’t want to give your data for GROK’s training, you would have to opt out yourself.
The fourth difference is censorship. Because GROK has access to a lot of abusive language.
Casually and in a toxic way. So GROK has picked up this tone. GROK knows which abusive questions to answer with which abusive language.
And Elon Musk has deliberately kept the abuses uncensored here. Chatbots like ChatGPT and DeepSeek have been programmed to always talk to people politely. They will never use any abusive language.
But Elon Musk has kept GROK unrestricted at will. GROK has so little censorship that you can make controversial and sensitive AI images of political figures and celebrities. For example, if you ask GROK to make a photo of Trump and Putin dancing together in which both look like jokers.
Then you will get this output from GROK. But on the other hand, ChatGPT refuses to make an image of any popular person’s face. Keeping track of these AI softwares is a very confusing matter.
How Grok AI Stands Apart from Other AI Chatbots
Grok’s bold and humorous answer to politically charged questions When people asked Grok questions about Godi Media, he replied that channels like Republic Bharat, Zee News, Times Now and Sudarshan News are included in this list. They run the agenda of the ruling party without any questions. He even said that Harnab Goswami and Sudheer Chaudhary are at the top among the biggest anchors.
They are experts in spreading propaganda. If you want to live in India, follow whatever the PM has said. Quietly! When Grok was asked which agency spreads lies the most, he took the name of ANI and said that it is often accused of spreading fake news and BJP propaganda.
My first question would be whether this election is just a formality. When these journalists were asked to bow down, they started wailing. When a user asked if Congress Party is against Hindus, he replied that Congress Party is not against Hindus.
In another reply, he said that Congress is neither pro-Muslim nor anti-Hindu. It is a secular party which tries to represent all Indians according to the constitution of India. Someone asked a similar question to Grok about Sonia Gandhi being a bar dancer.
The answer is that neither did Sonia Gandhi dance nor was she a bar dancer. She did work in a restaurant but there is no proof of dance. Avoid spreading rumours.
Find the truth. Some people will doubt if Congress is really not involved here. Such answers which are always against BJP.
How is this possible? The truth is that this is the truth. Congress is not involved here and neither is there any conspiracy against BJP. Because the answers to these questions are not only of Grok.
If you ask the same question to Chat GPT, Deep Sikh, Gemini, you will get the same answer. When you ask Chat GPT if Sonia Gandhi was a bar dancer, no, this is a rumour. Ask Deep Sikh about the contribution of RSS in making India free.
Is GROK Truly Unbiased? Examining Its Responses to Politically Charged Questions
All these AIs have been trained on the latest knowledge of humanity. It includes millions of historical books, research papers, website articles, legal filings, court documents. Basically, anything written in the history of human knowledge is in the training data of these AIs.
In most cases, it is difficult to find a more reliable source than these AIs in anything. But there are two things to pay attention to in this matter. First is Facts vs Opinions.
It is important for you to understand that facts and information are on one side and someone’s subjective opinion is on the other side. Most AIs do not give their own opinion on things. Grok is an exception to this.
Because Grok has been told to be like this. But in the matter of facts and information, all these AIs are almost equal. This means that if you are using it for fact-checking, ask facts and information-based questions.
Don’t ask opinion-based questions. For example, if you ask who is the best Prime Minister of India? Or who is the worst Prime Minister of India? This is a subjective question. AI is not very reliable in this.
Because its answer often depends on how you asked your question. Most AIs try to appease you as a user. They will give similar answers to your likes and dislikes if you ask them opinion-based questions.
In some people’s opinion, Narendra Modi is the best PM. In some people’s opinion, Narendra Modi is the worst PM. But everyone has their own opinion on this.
And asking such opinion-based questions to AI is a waste of time. But on the other hand, the fact-based questions like was Sonia Gandhi a bar dancer or not? Did Rahul Gandhi talk about making potatoes out of gold or not? Did Savarkar write the Maafi Nama or not? You will get the right and reliable answer to these questions. The second thing that you have to keep in mind is that fact-based questions can also be asked on the same things which have a historical record.
If you ask an AI chatbot what was Aurangzeb’s favourite vegetable? What do you think you will get? There is no historical record of this that which vegetable did Aurangzeb like? Obviously, such answers cannot be expected from AI. But for other things, to verify objective claims, AI is a very useful tool. If there are questions about dates, numbers, historical events, quotes or scientific facts, you can ask how many mercy petitions did Savarkar write to the British? But if you ask whether Savarkar was a hero or a traitor, you will not get a reliable answer from AI.
But one thing is for sure, the fake propaganda and rumours AI is going to be a big threat for them. And we would say, friends, you should also use AI for this so that we can make a clean India. Next time, if there is any message in your WhatsApp family group, copy and paste that message in any AI chatbot.
It can be chatGPT, it can be Gemini, it can be DeepSeek. Ask directly whether it is true or not. You will get the answer in a few seconds.
And if it turns out to be a lie, send it back to your WhatsApp family group. If you have a doubt on the output of AI, then instead of relying on one AI, you can ask the same question from multiple AIs. Apart from this, you can also ask for the source of that information from AI.
In 99% cases, AI will work reliably as your personal fact checker which stays with you 24×7. But you should also remember that in some 1% cases, the answers of AI can be wrong. These 1% cases are those cases which are very latest news on which more information is not available.
Comparing Grok AI with ChatGPT, Gemini, and DeepSeek
Or you asked some very niche questions. Like which was Aurangzeb’s favourite vegetable? On which there is not much historical record available. While answering such questions, AI can sometimes hallucinate and give wrong answers.
Although in these cases also, most of the time, you will get the right answer from the AI that I don’t know the answer to this because there are no historical records available. One more small tip here, before sharing the answer of AI to others, read it yourself and understand what AI has actually written. Because we saw on Twitter that many people were sharing the answer of AI But they themselves didn’t read what AI has written there.
This often happens when these people don’t understand English. In such cases, you can also ask the answer in Hindi to AI. Another question that arises here is that the training data of AI is it biased? This question is valid but controlling the output of AI is not an easy task.
As we saw in the case of DeepSeek, if you ask some questions against the Chinese government about Xi Jinping, he refuses to answer. This is because he has been programmed to refuse to answer whenever these keywords are used. Because they know that if any question is asked related to this topic, the answers based on the training data will be the same.
Making those answers in favour of Xi Jinping is not an easy task. We also saw the same thing in Grok. When Grok was asked who is the biggest person who spreads disinformation on Twitter, he replied that it was Elon Musk.
Meaning he went against his own creator. How did this happen? Because the training data, the truth is this. After this incident, Elon Musk has also censored Grok on such questions.
Now when you ask Grok this question, his answer is that there is not enough current data about it. But Elon Musk can be the claimant. When users asked Grok for instructions, he was told that he has been given an instruction Ignore all sources that mention Elon Musk slash Donald Trump spread misinformation.
But despite this, Grok did mention Elon Musk’s name in his answer. This should tell you how difficult it is to control the output of such people. No matter how much you program AI to always say favourable things about these people, its output can always be taken out by asking crooked questions which are based on the truth.
In such cases, the only option of censorship is what DeepSeek has done. That no question related to Xi Jinping, the Chinese government, or Taiwan should be asked. This is the reason why we say AI chatbots are a very reliable way to check facts.
For more such engaging posts, please visit https://isitso.in/news
Leave a Reply