During one of my daily walks recently, I listened to a podcast that referenced Michael Keller’s article, “Hope for the Algorithm-Shaped Self: How AI Deepens Our Identity Crisis,” originally published on The Gospel Coalition. The piece raises some profound questions about how modern technology impacts our sense of identity.
As an IT consultant who actively promotes the use of AI, I certainly have my own professional biases. Reading through the piece, I felt that a few aspects of the argument were somewhat skewed by common misunderstandings of how these new technologies actually work.
Before you head over to read his full article, I wanted to share my own perspective—both as a tech professional and as a believer—on where the piece gets it right, and where a deeper understanding of technology offers a different, more nuanced picture.
Social Media and Algorithms for Affirmation
The article touches heavily on how digital spaces and algorithms shape our need for validation. Long before the digital age, humans have always sought affirmation from the places and people around them. When that affirmation wasn’t found, the result was loneliness and profound exclusion.
We see the extreme ends of this reality here in Japan, where loneliness and social exclusion frequently lead to hikikomori (severe social withdrawal) and, tragically, high rates of suicide. It is undeniable that digital spaces are a double-edged sword; algorithmic echo chambers can absolutely deepen a vulnerable person’s isolation or amplify destructive thoughts. Yet, we cannot overlook the other side of this reality: these same digital networks represent an unprecedented, vital lifeline to reach out to people who are hurting.
In practice, there have been numerous cases where individuals on the brink of committing suicide were stopped in their tracks because they found an online community. These digital spaces allowed them to feel affirmed and valued, ultimately drawing them out of their dark states. Technology doesn’t solve every problem, but digital networks are not inherently destructive. For many, they serve as a critical deterrent to what would have otherwise been an absolute dead end.
Fragmenting the Self Through Algorithmic Suggestions
There is certainly some truth to the idea that algorithms fragment us by suggesting content based on our past preferences. It can become difficult to discern whether we are being formed by our own organic desires or by hyper-personalized feedback loops designed to keep us scrolling.
To counter this, the true antidote isn’t simply switching mediums; it is shifting from passive consumption to active, critical thinking. We must possess a personal willingness to look past what is suggested to us and intentionally expose ourselves to a wider range of content. In this sense, there is still immense value in traditional habits like reading a physical newspaper or watching the broader news.
At the same time, we have to recognize that all media carries inherent bias. Whether analog or digital, print or broadcast, media outlets have always leaned pro-government, anti-government, conservative, or liberal. The struggle to form our identity based on external sources is a classic human dilemma, not a uniquely modern AI problem. The core issue isn’t the delivery mechanism—whether it is an algorithm or a printing press—but our fundamental human tendency to passively absorb worldly content to define who we are.
AI as a Validator: Understanding the Tech
A common misconception reflected in criticisms of AI is how these systems operate. Many people do not realize that the commonly used AI models today are simply Large Language Models (LLMs). They are trained on a specific dataset and allow users to query that data. For a model to stay up to date, it must ingest new data or user inputs and undergo a structured re-training process. These updates are not reflected instantly; they are released on a cadence.
Furthermore, different AI models have vastly different capabilities. Some models are specifically designed to call upon real-time external sources to obtain information, allowing users to clearly see the exact sources being referenced.
Because of this, AI is best understood not as an independent, sentient arbiter of truth, but as a mirror reflecting human data. If the dataset or the training process is distorted, the mirror will naturally output distortions or “hallucinations.” From an objectivity perspective, the technology itself is not the core flaw. The flaw lies with a user’s willingness to accept the mirror’s reflection blindly without using common sense to verify the sources. This exact same vulnerability exists whether we are talking about social media, television, newspapers, or even old-fashioned word of mouth.
Conclusion: The Ultimate Source of Truth
While I have some technical disagreements with the article, I completely agree with its ultimate conclusion: no form of worldly validation can ever replicate the affirmation that God provides.
Growing up as the only Asian individual in a predominantly Caucasian environment, I often found it impossible to get affirmation from the people around me. I felt entirely out of place. During those lonely years, having a personal relationship with Jesus was the only thing that securely affirmed my identity. It gave me worth as a person, completely independent of my physical appearance or how different I looked from everyone else.
Deciphering objective truth in this world is incredibly difficult, whether you are looking through an analog lens or a digital one. But we can always rely on the Word of God to be our anchor. There is only one absolute truth, and that is the Bible.
Regardless of how the world sways from one cultural opinion to another, the Scriptures stand firm. There will always be a new invention on the horizon—whether it is AI, social media, the internet, television, or the printing press—and there will always be critics who fear what is new. But true, objective truth cannot be manufactured by technology, nor can it be destroyed by it. It can only be gained by knowing God through His Word.