2 minute read

Much of today’s `free’ or consumer internet is built on three principles:

  1. Involuntary user data exploitation: Recent debates around generative AI have returned attention to this.

  2. Novel manipulation of users by malicious actors through personal chat: WhatsApp is used as a political weapon to broadcast misinformation during elections, especially in developing economies. It was central to disinformation campaigns during elections in Brazil El-Masri, Azza, Riedl, Martin J, and Woolley, Samuel. “Audio misinformation on WhatsApp: A case study from Lebanon.” Harvard Kennedy School (HKS) Misinformation Review, 2022. , IndiaBanaji, Shakuntala, Bhat, Ramnath, Agarwal, Anushi, Passanha, Nihal, and Sadhana Pravin, Mukti. “WhatsApp vigilantes: An exploration of citizen reception and circulation of WhatsApp misinformation linked to mob violence in India.” Department of Media and Communications, London School of Economics and Political Science, London, UK, 2019. , and LebanonRecuero, Raquel, Soares, Felipe, and Vinhas, Otávio. “Discursive strategies for disinformation on WhatsApp and Twitter during the 2018 Brazilian presidential election.” First Monday, 2021. .

  3. Social media companies incentivize poor online behaviors: X and Reddit allow users to publish posts rapidly; some are insightful and informative while others lack taste, to put it mildly, leading to polarizing toxic content.

Proposal for new mechanics of online publication could ameliorate today’s consumer internet

I have been thinking a lot about the fundamental mechanics of online social platforms and how they can be designed to be less exploitative/manipulative and offer more respectful experiences on the internet. Specifically, I have been thinking about how publication should work on the internet and how this should be facilitated by personal reflection. This would work better, in my opinion, than online public shaming/arguments on comments threads of existing platforms.

So, my idea is to re-imagine modern-day social media and personal messaging systems to introduce a new way to publish that allows content publication/broadcasting only using content that exists within a chat in your inbox. If this sounds unclear, think of having a button on WhatsApp or Gmail that converts a set of chat messages/emails/drafts between you and some friends into a blog or project proposal or white paper. In addition to human collaborations making the content appropriate for public sharing, an on-device AI agent (not on the cloud) enhancing/improving your pre-publication work, exposing biases in your thinking (think of this as an AI agent commenting in your GDocs). And then a button allows you to publish this content (or warns you against publishing). The on-device AI preserves privacy and recent developments show that LLMs can work on current personal computing hardware. A cloud-based agent could also be used for a second stage of monitoring; this would not violate user data privacy because the user’s intent with posting is, after all, to have the data examined.

I believe this introduction of some friction in the publication process that encourages reflection is one step towards driving individual users towards a society that’s fairer and more respectful.

Some personal progress on making such a platform

In this spirit, I spent some free time working on an inbox-to-publication platform using audio/speech as the communication modality. This choice was informed by my Indian heritage, where I feel spoken literacy is higher than written literacy which is being exploited as I mentioned earlier. Click here to see an (out-of-date) demo of an iOS called Howler from 2022.