Beware! AI can wreak havoc ahead of next polls & after

Dear politicians of all hues please don’t take this lightly. Artificial Intelligence or AI that is being given a big push across the board can turn out to be a double-edged weapon.

I will not hesitate even calling it an unguided missile as it will cause destruction all around. You may not see buildings crashing and airports being blown up, but the AI missiles have the potential to destroy the thought process and eventually, democracy.

Is it going to be the enemy of democracy? Yes, if it is not handled well and if the proper ground rules, surveillance systems and strict code of conduct are set in place. Aren’t we witnessing the IT cells working overtime to downplay, defame and destroy political opponents using the social media tools? AI impact will be even worse.

Sample this. The meme on PM Modi dancing on India landing on the moon has earned quite a few eyeballs – positive and negative. Podcast journalists have even questioned the big media house that put out the meme for exhibiting its ‘uncalled for obeyance’ to the ruling dispensation! So far so good, if it stops here. But the portending danger is that the cartoon character created for this meme can be copied using the AI tools. Video shorts showing the character in different roles and with different messaging. The narrative can be skewed to suit one’s political ends.

We have just seen the impact of a Gabbar meme, created by the Congress, with NaMo as Gabbar Singh blasting the trio of Amit Shah, Nadda and Rajnath Singh with tomatoes on their failure to check Rahul Gandhi! The clip posted on X platform (formerly twitter) has earned 890.7K Views, 31.4K Likes, 8,669 Reposts, 477 Quotes as per the count on last Sunday morning. Earlier, we have seen a series of memes with BJP leaders dancing to the tune of Dil Ding Dong Ding Bole as they purportedly ‘celebrated’ the return of their star Pracharak, Rahul to the Lok Sabha.

Then we have pro-BJP (rather pro-Modi) TV channels deriding Rahul Gandhi on some pretext or the other and countless standup comedians and podcasters ridiculing Modi on anything and everything that he does.

These are all fine as long as they do not cross the Lakshman Rekha of pure humour. But with the use of AI tools all the examples that I described can be tweaked into possible weapons to spread disinformation. IT tools can be misused to hack into political opponents’ social media handles and websites and cash in on the already polarised atmosphere of hatred.

Here, I do not want to give any examples that may be used as ideas for the IT cells. But can definitely hazard a guess on the scenario in the not-too-distant future.

Today, any content can be created and edited on ChatGPT as per one’s preferences. For instance, if you ask for negative points of Modi and Rahul Gandhi, ChatGPT gives at least eight attributes for each and one has the option of keeping on asking for regenerating more versions. Such tools are definitely useful for harmless, genuine writers and researchers but my worry is the certainty of their misuse.

One may argue that nuclear energy can be used for making bombs as well as generating power. That’s why the international community and the UN have the checks and balances on the Nuclear front.

So, what about AI which can be used for positive and negative purposes?

As regards the checks and balances in social media, allegations are that they are being used to block political opponents’ or critics’ accounts. Can AI not meet a similar fate? This has to be discussed thoroughly before we all go on an overdrive with AI.

One good check that I experienced with ChatGPT is that you can use it only through google verified accounts. It may not be difficult for google to track down any criminal misuses. But what happens if the ChatGPT content is edited to suit one’s goals and used for causing mischief? And what happens if the content is used as the base propaganda material through the traditional print or electronic media and then extrapolated via the digital media?

My attempt here is to play the devil’s advocate and not propagate devilish ideas!

I firmly believe that the government must devise an inclusive policy on the guidelines and framework for using AI tools and ensuring checks and balances. It is not enough to get the official, one-sided expert opinion. It is important to involve crooks as well.

Police often learn the criminals’ modus operandi by obtaining the description from the latter. Remember Ajmal Kasab who spilt all the beans of Pak conspiracy and helped Mumbai police map the Karachi-to-Kutch-to-Mumbai route of the attackers’ squad?

Also, it is important to ask people, influencers to list out the ways of misusing the tools and plan to checkmate them with the help of cybercops from across India.

But the BJP government has the record of coming out with policies without involving the public opinion. Moreover, the party is in a hurry to push through the one-nation-one-election policy to strengthen its double engine doctrine.

The hopes of a united opposition have increased with the leaders of INDIA alliance speaking in a united voice at their Mumbai meeting last week. The alliance has appointed committees on various subjects such as media freedom, social media and joint action plan. This alliance must also critically examine all social media aspects including the AI.

Here, we discussed the impact of AI only from the political perspective. AI applications are enormous from industry to medicine, hospitals to hospitality. The very thought is mind boggling.

Let us all put our brains together to evolve checks and balances and leave full scope for course correction. What happens if…should be a permanent point of discussion. Who knows, the devils can come out with a set of 10 IT Commandments!

(The columnist is a Mumbai-based independent media veteran, running websites and a youtube channel known for his thought-provoking messaging.)