How AI Influencers Are Making It Hard To Trust What We See Online

In a world that is driven by artificial intelligence, the demarcation between real and virtual is disappearing faster than we realise. You can just scroll through Instagram for a couple of minutes, and chances are that you will come across a perfect face, a convincing voice, and a believable story—only to figure out that none of it is real. From popular virtual influencers like Lil Miquela that collaborated with brands like Dior, Prada, and Samsung to India’s own Kyra landing partnerships with boAt and Amazon Prime Video, synthetic personalities have strongly entered the mainstream. But the real problem arises when AI-generated influencers and deepfakes start to take control of how and what we trust online. As per cybersecurity expert Dr. Anil

Rachamalla, Vice President at FourthSquare and co-founder of the Council for Digital Safety & Wellbeing, the real danger is not in the development of AI but rather how it is being weaponised. The Problem: How Digital Influence Is Turning Into DeceptionAI influencers are no longer just a novelty. They are training a wider audience to accept that synthetic faces are normal. And Rachmalla argues that this drastic shift is deeply alarming. He stressed, “People are getting used to seeing these faces. The familiarity is increasing, but the truth behind them is not.”The growing acceptance for virtual influencers is being exploited at a large scale. Scammers are layering scams on top of hyper-realistic AI-generated content, such as investment pitches, creating fake endorsements and setting up emotional narratives that are difficult to identify. Deepfakes of celebrities such as Amitabh Bachchan, Shahrukh Khan, and Rashmika Mandanna have been circulated online, often used to promote fake products and manipulate the larger public sentiment. Some of these cases even include political figures and financial authorities to deceive people to lend credibility to scams. The result is what Rachamalla describes as a ‘familiar face but unfamiliar truth,' where audiences trust what they see, even when it is entirely misleading.Use Cases: From Emotional Manipulation To Financial FraudsAccording to Rachamalla, the misuse of AI influencers and deepfakes spans several categories listed below:Financial ScamsFraudsters are creating believable AI videos of public figures and popular influencers claiming huge returns from investments. These often direct users into WhatsApp or Telegram groups, where they are gradually manipulated into transferring money, with minor or no chance of recovering their deposits. Voice Clone FraudsPopular AI tools like ElevenLabs are being used to replicate your family member’s voice within minutes. Victims receive fake calls from what may sound like their family members, leading to panic-driven financial transfers.Sextortion And BlackmailUnlike earlier methods needing compromising media, AI now lets criminals generate explicit content using just an image. Victims are then blackmailed, even though the content is purely AI-generated. Emotional Hooks And Engagement FarmingDramatic videos are created showcasing politicians and celebrities to capture attention online and increase watch time to monetise engagement rather than directly scam anyone. Political ProvocationRachamalla highlights how emerging deepfake cases are used to create tensions between communities and religious groups, creating communal tensions. He notes that even the entertainment industry is not behind with such experiments. The cybersecurity expert cited an example of a Spanish show, Falso Amor. It manipulates AI-generated intimacy scenarios to provoke emotional reactions, only to reveal later that the clip was not real.

How The Problem Is Already MainstreamThis is not about scrolling anymore. Rachamalla estimates that nine out of 10 Indians are likely to witness synthetic content during their random social media surfing. In merely 10 minutes of scrolling, audiences will see some sorts of misleading and manipulated content online. He believes that not even one in a thousand people get their money back, asserting how organised and normalised these scams have turned out. The reach of these fraudulent and deceiving people has no borders now. Unlike traditional scams defining a demographic, AI-driven deception reaches any age group, geography, and literacy level where the only condition is that the victim should own a smartphone. How Can You Trust What You See Online?The most ignored and dangerous impact of these frauds is psychological. AI influencers are not just used for deceiving people; they are conditioning them. As users monotonously consume AI-generated content, their ability to question authenticity weakens, asserted Rachamalla. This has led to increased misinformation, emotional attachment involving romantic feelings for virtual creators, and a reduced instinct to verify content before making choices about it.Rachamalla warned that the audience is not getting trained to detect AI; instead, the AI is training the audience to accept it.

What Are The Solutions?Despite the scale of the problem, Rachamalla highlights how audiences can take back control. Spot Audio And Video Red FlagsAI videos often have imperfect lip sync, unnatural blinking patterns, unnecessarily polished or waxy skin textures, and even inconsistent jewelry or clothing details.Check The ContentIf you spot a celebrity or a politician with an unusual accent, request, or message, do not believe it immediately. Pause and check the source of the video along with its context to fact-check. Free AI tools like InVID or Gemini SynthID are available to help spot manipulated media. Report IssuesRachamalla highlighted how victims can report issues with India's cybercrime helpline (1930) or via the official portal of cybercrime.gov.in.As Rachamalla suggests, the real battle is not against AI influencers; it is against the quiet erosion of trust itself. The AI-generated content will become more sophisticated and integrated into marketing day by day, but the core idea is to perceive the right context.