The dark side of AI: When 'seeing' is not 'believing'

Narayan Devanathan, group chief strategic advisor, Dentsu

 

The language and conversations around it, especially from lawmakers (whether Rajeev Chandrasekhar or Joe Biden) and opinion makers (like Amitabh Bachchan), is quite telling. To treat it as a legal/ technological problem primarily. And it is right actually, because as humans have repeatedly demonstrated, we’re incapable of using dangerous technology responsibly. Remember guns? The gun-loving lobby will cry hoarse about the absolute right to bear arms and that 'guns don’t kill people, people kill people, so what’s the point in outlawing guns?'

 

I don’t see any of them saying 'deepfakes don’t ruin people, people ruin people, so what’s the point in outlawing deepfakes?' And while the likes of Joe Biden may not have shown the gumption to pass strict gun laws, I’m glad to see that they’re at least jumping onto the bandwagon early in this case. There’s an urban legend around the drafting of the U.S. Constitution that the founding fathers started with one principle to put in checks and balances: people cannot be trusted, especially with power. That holds true in the case of (mis)use of AI and deepfakes. So yes, bring on the strict regulatory frameworks already.

 

There’s an old and famous anodyne that corporates use to try and 'humanise' themselves: companies are people too. I’m going to use it in this case to equate the (online) platforms with people with regard to the same character flaw: they can’t be trusted, especially with power. To expect platforms to take the onus on themselves to curb the spread of content is like arming them with a lot more matches and flammable liquid and asking them to make sure the fire doesn’t spread. The platforms have shown themselves to be incapable or unwilling to curb the massive spread of fake news. I don’t hold out much hope about their self and other-regulating capabilities when it comes to the (mis)use of AI and deepfakes either.

 

Samir Asher, co-founder and COO, Tonic Worldwide 

This is just the tip of the iceberg, as the technology gets mainstream, bad actors will use it to their advantage. It requires a holistic approach where both the regulatory framework and the platforms will have to step in and create barriers for the misuse of the technology. In my opinion, the platforms on their part will develop systems that will be able to detect deepfake and flag off the content or add a note so the user is aware of its authenticity and also create efficient processes for user reporting and taking down manipulated content. X, Facebook and Instagram, etc, can easily develop a tech that flags off if the content is deepfake.

 

Himanshu Arora, co-founder, Social Panga & The Yellow Shutter

 

Similar to the corrective measures and efforts initiated by regulatory bodies and brands to address financial fraud in the BFSI industry, there is a growing recognition of the need for proactive strategies and countermeasures to tackle the challenges posed by deepfake technology in the content sector. This issue extends beyond mere inconvenience, as the manipulation of audio and visual content raises significant concerns about misinformation and deceptive practices.

 

Furthermore, it is crucial for these platforms (social media) to engage in educational initiatives, raising awareness among users about the existence and potential risks of deepfake content. By doing so, social media platforms empower users to be more discerning consumers of information.