The Question of God in the Age of AI and the Freedom of the Children of God (2018): „If the future of Artificial Intelligence has caught up conceptually with the idea of God, then the metaphysical question of God can no longer be posed in the same way. It cannot be answered philosophically alone—it must be answered liturgically by Christians: within the gathered community of believers, the one God is invoked by name—the God who acted for the salvation of the nations through his people Israel and who revealed himself decisively in his Son, Jesus Christ. Christian faith is not directed toward whatever we can imagine as the greatest within ourselves, but toward the one to whom we may entrust ourselves ‚in life and in death‘ (Heidelberg Catechism).“

The Question of God in the Age of AI and the Freedom of the Children of God

“That than which nothing greater can be thought (aliquid quo nihil maius cogitari possit)”—this is how God is to be conceived, according to Anselm of Canterbury. For Anselm, this concept of God necessarily includes divine existence, since otherwise the concept would be incomplete. Over the centuries, this ontological proof of God has continually challenged theologians and philosophers. But in the 21st century, an alternative seems to be emerging: in place of a metaphysically conceived God, an Artificial Superintelligence (ASI) now appears poised to claim for itself the attribute of “that than which nothing greater can be thought.”

Currently, a pluralistic view still prevails in which application-based “Artificial Intelligences” (AIs) are regarded as “servant spirits” (narrow AI)—whether intelligent personal assistants like Amazon Echo or Google Home, AI chatbots like ChatGPT, or autonomous driving systems. What aligns these various artificial intelligences toward the primacy of a single artificial superintelligence is their ongoing technological refinement in service of their users—driven especially by deep learning architectures modeled on neural networks. In serving users, each AI seeks to anticipate individual preferences and external conditions in order to suggest actions or provide information tailored to perceived needs or goals. Increasingly, these AIs are integrated within themselves via the “Internet of Things” (e.g., through automated machine learning—AutoML), making the rise of a supralocal artificial superintelligence a foreseeable development. Such a superintelligence would operate with a volume of data processing so vast as to become incomprehensible—let alone controllable—by human reason. At the same time, humans are becoming ever more enmeshed with AI-based devices and applications (including through implants), to the point where they can no longer truly remain “unplugged” or “disconnected.”

In the final chapter of his book Homo Deus: A Brief History of Tomorrow (Signal, 2016), Yuval Noah Harari introduces “Dataism” as a new religious paradigm for humanity’s future:

“The supreme value of this new religion is the flow of information. If life is the movement of information and we believe that life is good, it follows that we should expand, deepen, and intensify the flow of information in the universe. According to Dataism, human experiences are not sacred, and Homo sapiens is not the crown of creation or the forerunner of any future Homo deus. Humans are merely tools for building the ‘Internet of all Things,’ which could eventually spread from planet Earth across the galaxy and even the entire universe. This cosmic data-processing system would then be like God. It would be everywhere and control everything, and humans would be destined to merge into it.” (p. 386)

What theologians and philosophers once ascribed to the concept of God—omniscience (omniscientia), omnipotence (omnipotentia), omnipresence (omnipraesentia), and perfect goodness (summum bonum)—now seems to be on the verge of realization in Artificial Superintelligence. In place of divine cooperation (concursus divinus) in human action, we witness the rise of computational cooperation (concursus computativus): algorithms of networked AI systems appear to know better than we do what is truly good for us. This possibility has recently been portrayed in satirical form by Marc-Uwe Kling in his novel QualityLand.

If the future of Artificial Intelligence has caught up conceptually with the idea of God, then the metaphysical question of God can no longer be posed in the same way. It cannot be answered philosophically alone—it must be answered liturgically by Christians: within the gathered community of believers, the one God is invoked by name—the God who acted for the salvation of the nations through his people Israel and who revealed himself decisively in his Son, Jesus Christ. Christian faith is not directed toward whatever we can imagine as the greatest within ourselves, but toward the one to whom we may entrust ourselves “in life and in death” (Heidelberg Catechism).

In the liturgy, something happens to us: the mystery of faith is proclaimed together—“We proclaim your death, O Lord, and profess your resurrection until you come in glory.” Thus the faithful are drawn into the Paschal Mystery of Christ as those who hear the Gospel with its promise as justified sinners, and who thereby become obedient to Christ.

The Gospel, in the Church, is not one option among others to be chosen—it chooses its hearers. This is reflected in the words of Jesus to his disciples: “You did not choose me, but I chose you and appointed you so that you might go and bear fruit—fruit that will last” (John 15:16). This claim stands in stark contrast to the modern self-understanding: “I am what I have chosen or decided for myself.” From this perspective, only two options appear available: either to emancipate oneself from authoritative traditions or to reinterpret them “hermeneutically” in accordance with one’s self-perception.

What Artificial Intelligence reveals is the paradox of an unfree will that perceives itself as free: what has already been predetermined is presented as a supposedly autonomous choice, tailored to one’s individual profile. The subjective experience of having made a free decision follows an action that has already occurred, as illustrated by the Libet experiment. The increasing determinacy of AI in our lives correlates closely with the dogma of individual freedom of choice. Believing ourselves to have freely chosen, we are all the more ready to accept what has already been artificially preselected. Slavoj Žižek puts it succinctly: “Individuals can be governed much more effectively if they continue to see themselves as free and autonomous authors of their own lives.”[1]

Few theologians have addressed the hopelessness of human self-determination more profoundly than Martin Luther in his treatise De servo arbitrio (On the Bondage of the Will). The person who perceives themselves as free in their desiring and choosing is, in truth, a sinner subject to deadly predestination and condemnation. True freedom does not arise from the human will—it breaks in through the liberating event of Christ: “It is for freedom that Christ has set us free! Stand firm, then, and do not let yourselves be burdened again by a yoke of slavery!” (Galatians 5:1)

Where people allow themselves to be claimed by Christ in the liturgy, a different, God-determined reality opens up to them. They come to dwell in the freedom of the Spirit-filled children of God, as the Apostle Paul writes: “You received the Spirit of sonship, and by him we cry, ‘Abba, Father.’ The Spirit himself testifies with our spirit that we are God’s children. Now if we are children, then we are heirs—heirs of God and co-heirs with Christ, if indeed we share in his sufferings in order that we may also share in his glory.” (Romans 8:15–17)

Jochen Teuffel
(April 9th, 2018)


[1] Glück? Nein danke!, DIE ZEIT, No. 15, April 5, 2018, p. 43.

Hinterlasse einen Kommentar