Generative AI Comes to Social Media: The Good, the Bad, and the Uncertain

Source: Snapchat

Snapchat made headlines earlier this week by announcing the launch of My AI, a new ChatGPT-powered chatbox, which comes through ChatGPT’s newly announced API.

Part of Snapchat+, the platform’s growing subscription service, My AI serves as a sidekick that users can name, customize, and of course, interact with. My AI is similar to the native ChatGPT platform, but with more restrictions to comply with safety and trust guidelines.

Snapchat Isn’t Alone

Around the same time, Mark Zuckerberg announced that Meta was in the process of constructing a team dedicated to building generative AI across its experiences. It plans to start with creative and expressive tools and eventually move to “AI personas.”

Just a few days later, Neal Mohan, YouTube’s new lead, penned a letter to creators about YouTube’s priorities for 2023. He revealed that the platform too would invest in generative AI. He mentioned the platform would build features that will reinvent video and help creators to story-tell and increase their video production value. Potential features allow creators to virtually switch outfits in videos and create unique film settings.

It’s also reported that Elon Musk is looking to build an AI lab in order to build his own ChatGPT rival, which likely would be integrated into Twitter. Plus, earlier this month, Reddit said that it's exploring AI chatbot technologies as well.

Why Social Media Platforms Are Investing in Generative AI

Social media platforms bringing generative AI to their apps is much more than checking the box. Generative AI has real potential to help them drive business in a number of ways:

• User Engagement: By providing creators with AI-powered tools, social media platforms can make content creation easier. Not only can this help its creators to create and share content, but it also can lower the barrier to entry for everyday social media users to become creators. Both of these scenarios can help drive user engagement, a key metric for many platforms. Because the use of, for example, picture editing tools can lead to significant time efficiencies, which means people can produce high-quality work without sacrificing their time resources.

• First-Party Data: Social media platforms can gather a wealth of first-party data from generative AI like chatbots. As users interact with these conversational messaging tools over time, platforms obtain a lot of information about them, which can be used for advertising purposes. This includes displaying relevant ads within chatbot experiences and serving ads across other experiences and formats. Apple's privacy changes have made it challenging for social media platforms to offer precise targeting for advertisers, and many advertisers are reducing their spending due to this. First-party data can help platforms provide advertisers with more accurate targeting, which could lead to increased ad budgets.

• Subscription Revenue: As Snapchat has shown, adding generative AI features can be a way for social platforms to get dollars directly from app users. Snapchat's subscription offering was already strong, with over 2.5 million subscribers. The addition of My AI should convince more to sign up for it. With platforms diving deeper into paid features, putting in-app generative AI tools behind a subscription could make it easier for users to pay for their social media usage.

With the good comes the bad. Social media platforms getting into generative AI raises some concerns:

• Copyright Infringement: By using preexisting content to create new images, videos, texts, or other formats, generative AI may infringe on the intellectual property rights of creators. Recently, social platforms have made efforts to help creators get proper credit and attribution for their work through new tools and initiatives. If this is still a priority for them, they will need to consider how to think about this in the context of generative AI, which to date often doesn’t credit the source(s) of work being used to create something “new.” Another thing that could be an issue is if creators use generative AI for branded content. For the most part, creators and brands are restricted from using creative effects, commercial music, and more unless they have permission. This likely will be the same case for content that is built using generative AI. If creators become too reliant on using these tools to produce content, brands put themselves at risk when working with these creators.

• Mental-Wellbeing & Misinformation: Another concern is chatbots, such as Snapchat's My AI, being positioned as digital friends. If people engage with chatbots more than real people, it can create a false sense of connection and make people feel more isolated. Social media was created to help people connect more, but with the advance of technology, it has resulted in the opposite. Platforms are leveraging technologies to maximize time spent in-app. There is also potential for misinformation, which is already a big issue across social media with human-generated content.

• Data Privacy: Data privacy is also a significant concern with the use of generative AI. Users already provide vast amounts of personal and professional data in exchange for platform access. The addition of generative AI would only increase the amount of data controlled by social media platforms, further eroding users' control over their personal information. Given the desire for greater control over personal data, the incorporation of generative AI may be counterproductive.

Ever since generative AI got on my radar in October, I've had a similar thought about what it means not just for the creator economy but for everything going forward. Generative AI will follow a similar journey as non-fungible tokens (NFTs), where creators, brands, and VCs see it as the next growth hack and cash grab, but it comes with equal opportunities and controversy.

Overall, I believe the emergence of generative AI is a positive development that will significantly impact our lives going forward. However, it is crucial for platforms, companies, and everyone in between to consider how to use and integrate it responsibly.