AI Worship Music: What Solomon Ray’s Charts Exposed

AI-generated worship music can top the charts, but it cannot lead worship — because worship is an offering rendered by living worshipers, and a generated track has no one behind it doing the offering. When the AI “artist” Solomon Ray reached No. 1 on Christian charts in late 2025, the scandal was not that a machine made plausible worship music; it was that so much human worship music had become plausible enough for a machine to imitate. The faithful response is neither panic nor ban, but a recovered theology of musical craft — and for East Africa’s surging gospel scene, a once-in-a-generation invitation to write what no model can.

Key Takeaways

• “Solomon Ray,” an AI-generated gospel artist created by Mississippi rapper Topher (Christopher Townsend), reached No. 1 on iTunes’ Top Christian & Gospel Albums chart, with two songs topping Billboard’s Gospel Digital Song Sales chart in late 2025 (1)(2).
• The backlash was immediate and industry-wide: Dove Award winner Forrest Frank objected that “AI does not have the Holy Spirit inside of it,” while the real Solomon Ray — a worship leader in Montana — said God wants “costly worship,” not machine-crafted songs (3)(4).
• Christianity Today’s verdict gave this article its title: the current No. 1 Christian artist has no soul — no testimony, no congregation, no life behind the lyrics (4).
• Critics across Baptist and mainline publications converged on a deeper diagnosis: AI worship charted because much human-written worship had already become formulaic — the machine exposed the template (5)(6).
• The theological core: worship is covenant offering, and offering requires witness, congregation, and sacrifice — three things generation cannot supply. David’s rule still governs: “I will not offer to the LORD that which costs me nothing” (2 Samuel 24:24).
• The stakes are commercial as well as doctrinal in East Africa: gospel streams in Sub-Saharan Africa grew roughly 50% year-on-year, with Kenya up 30% and gospel ranking among the country’s most-streamed genres — a boom AI catalogs will target (7).

What Actually Happened with Solomon Ray?

The facts are stranger than the headlines. In November 2025, an AI-generated gospel singer named Solomon Ray — invented by Christopher “Topher” Townsend, a Mississippi-based conservative rapper and Air Force veteran — climbed to the top of the iTunes Top 100 Christian and Gospel Albums chart with the EP Faithful Soul, while the tracks “Find Your Rest” and “Goodbye Temptation” took the top two positions on Billboard’s Gospel Digital Song Sales chart (1)(2). Billboard’s reporting traced the pipeline: AI generation tools, a fictional persona, and distribution channels that treated the output like any other release (2).

The industry’s reaction split along a fault line worth examining. Working artists objected on grounds both spiritual and economic. Forrest Frank, a Dove Award-winning CCM artist, put the spiritual objection in one sentence: “AI does not have the Holy Spirit inside of it” (3). Human musicians also noted the unfairness of competing for chart positions and playlist slots against an entity with no studio costs, no band to pay, and no life to live. Meanwhile, in a detail almost too apt for a sermon illustration, there turned out to be a real Solomon Ray — a flesh-and-blood worship leader in Montana who has served with Elevation Worship. His response to his synthetic namesake cut to the center: it saddened him to see people “hoodwinked by AI,” because God desires costly worship rather than songs crafted by machines (4).

Topher defended the project as “an extension of my creativity,” arguing that “God can use any vehicle to reach people” (8). Set aside the motive question — Christianity Today, Religion Unplugged, and Baptist Press all noted the unresolved issues underneath: Suno and similar generation tools face major-label lawsuits alleging their training data was ripped from copyrighted recordings, meaning AI worship tracks may be built, without consent or compensation, on the sampled labor of the very human artists they now displace (1)(5). Worship strategist Kenny Lamm urged “slower, better-informed discernment rather than blanket fear” — the church has absorbed technology from the printing press to the podcast — while holding that AI can assist the craft but cannot replace the heart and witness of human worshipers (9).

That is the right instinct. But it needs a theology under it, because “discernment” without criteria collapses into taste. So: what exactly does worship require that generation cannot supply?

What Does Worship Require That AI Cannot Supply?

Start from what worship is. In the Reformed understanding, congregational worship is covenant renewal: the assembly of God’s people, gathered at his call, responding to his Word with their own confession, praise, and offering. The songs are not ambience for a spiritual experience; they are the assembly’s speech — “addressing one another in psalms and hymns and spiritual songs, singing and making melody to the Lord with your heart” (Ephesians 5:19). Three requirements follow, and each names something a generated track structurally lacks.

First, worship requires a witness. A worship song is testimony set to music — “He drew me up from the pit… and put a new song in my mouth” (Psalm 40:2–3). The song’s authority is not only its doctrinal accuracy but the life behind it: someone was actually delivered, actually grieved, actually trusted God in the dark and wrote it down. This is why the church sings “It Is Well” differently after learning Spafford wrote it over the sea that took his daughters. An AI track can assemble the vocabulary of testimony — rest, temptation, grace — but there is no one to whom any of it happened. Christianity Today’s framing was theologically exact: the No. 1 artist has no soul (4). Not “low quality.” No witness. The category of testimony does not apply to it, the way the category of marriage does not apply to a toaster.

Second, worship requires a congregation. Songs for the church are written to be owned — singable by an assembly, in its language, bearing its actual griefs and hopes before God. This is where the critics’ deeper diagnosis lands. Baptist News Global argued that the AI controversy exposed how formulaic and interchangeable much commercial worship had already become (6); Baptist Press asked plainly whether AI-generated worship generates worship at all (5). A machine could chart in the genre because the genre had drifted toward product: songs written for playlists rather than for particular peoples. The healthiest worship cultures in the world — and East Africa’s polyrhythmic, multilingual, call-and-response traditions are among them — are the least imitable, precisely because their songs grow out of specific congregations. Generic input, generic output: the model can only flatten toward the average of its training data, and the average has no village, no language, no funeral, no harvest.

Third, worship requires sacrifice. Here the real Solomon Ray handed the church its text. When David was offered a free threshing floor for his altar, he refused: “I will not offer burnt offerings to the LORD my God that cost me nothing” (2 Samuel 24:24). Offering, by definition, costs the offerer — and musical offering costs years of practice, hours of rehearsal, the vulnerability of standing before God’s people with your actual voice. Generation is the engineering of costlessness; that is its entire value proposition. Which means a fully generated “worship offering” is a contradiction in terms: an altar built to eliminate the sacrifice.

These three requirements give us the framework I teach worship teams, and I give it a name so it can travel: the Araunah Test, after the threshing floor David refused to take for free.

The Araunah Test: Three Questions for Any Song, However It Was Made

1. Witness: Is there a worshiper behind this song — someone whose testimony, repentance, or hope it carries — who can stand before the congregation answerable for it?
2. Congregation: Can this assembly own these words as its own speech to God, in its language and its life?
3. Sacrifice: Did this offering cost the offerer something — labor, craft, vulnerability — or was costlessness the point?

A song must pass all three to be led in worship. Notice what the test does not ask: whether AI touched the process. That is deliberate, and it is where a working theology of craft must be honest in both directions.

Where Is the Faithful Line for AI in Musical Craft?

Musicians have always used tools, and some of those tools now have AI in them. A theology of craft that cannot distinguish a notation aid from a generated artist will be ignored by every working musician under thirty — rightly.

The line follows the same architecture I have argued for preaching: leverage for the work, never substitution for the witness. Run it through the worship workflow:

AI may serve the scaffolding. Chord-chart transcription, stem separation for rehearsal tracks, arrangement suggestions a human arranger judges, mixing assistance, first-draft translations of a finished song into a congregation’s second or third language (with human poets refining them), searchable archives of a church’s own repertoire. None of this touches witness, congregation, or sacrifice; all of it returns hours to the craft. The builder’s question — how believers make tools as sub-creators under the Creator — applies here with full force: these are legitimate things to build and to use.

AI must not supply the offering. Generated lyrics presented as testimony, generated vocals presented as a worship leader, fully generated tracks presented — or quietly playlisted — as the praise of people who never praised. And one bright procedural rule binds the gray zone: disclosure. Religion Unplugged’s survey of church leaders found near-unanimity that labeling and transparency are non-negotiable (1). A congregation that discovers it has been singing along to a fabricated witness loses something it cannot easily get back, in exactly the way a congregation discovering machine-drafted sermons does.

Held together: the Araunah Test governs what may be offered; the leverage-substitution line governs how it gets made; disclosure governs the space between.

Why Is This East Africa’s Moment Rather Than East Africa’s Threat?

Now the part the chart-watchers are missing, and the reason this article is hopeful rather than defensive.

East Africa is in the middle of a gospel-music boom that is the envy of the global industry. Gospel streams across Sub-Saharan Africa grew roughly 50 percent year-on-year into 2025; Kenya alone logged a 30 percent rise, with gospel ranked among its top-ten streamed genres; the Kenya–Uganda–Tanzania–DRC live circuit generates an estimated $20 million in annual tour revenue; and Ugandan gospel artists are now mounting productions budgeted in the hundreds of millions of shillings (7)(10). Listeners are actively rebalancing toward African artists singing in their own languages and worship styles (7). This is a pillar of the region’s creative economy and a leading edge of the faith economy as an investable sector — jobs, studios, royalties, and discipleship braided together.

Into that boom, the cheap catalog is coming. The economics guarantee it: if generated gospel can chart in Nashville, generated “Swahili worship” playlists assembled by no one, for no congregation, at near-zero cost will flood the platforms next, skimming streams from the very artists driving the boom.

But look at the matchup through the Araunah Test and the African advantage becomes obvious. What is the model’s training data for a Luo funeral song? For the way a Kampala congregation bends a chorus around a testimony shouted from the third row? For Acholi call-and-response, or a Kinyarwanda lament composed out of a history no other nation carries? The thinner a tradition’s witness, the easier it is to generate; the thicker, the harder. East African worship is among the thickest, most local, most congregational music on earth — the global style least replaceable by a machine averaging the internet.

So the strategy is not a ban; it is abundance. Write more. Write deeper. Write local. Churches should commission songs from their own musicians the way they once commissioned buildings — for the harvest, for the burial, for the baptism, in the mother tongue. Worship leaders should use every scaffolding tool AI offers — charts, translation drafts, rehearsal stems — to multiply output of the thing only they can supply: witnessed, congregational, costly praise. Labels and platforms in Nairobi, Kampala, and Dar es Salaam should adopt disclosure standards for generated content before the flood, not after. And the region’s church-tech builders should treat “tools for human worship makers” as a product category — part of the larger case that Africa can lead the church’s AI era rather than import its mistakes.

The charts exposed something in late 2025, but not what most commentary claimed. They did not expose AI’s power; generation is just statistics with a microphone. They exposed how much of our worship had already gone generic enough to be generated — and, by contrast, how irreplaceable the real thing is. A machine can make a plausible song about rest. It cannot drive to the hospital, sit with the dying, and come back Sunday with a song that costs something. That is the musician’s office. No model is coming for it — and the congregations of East Africa, singing their own new songs in their own tongues, are about to prove it to the whole church.

FAQ

Who is Solomon Ray and why did the AI version cause controversy?
Solomon Ray is an AI-generated gospel “artist” created by Mississippi rapper Topher; its EP topped iTunes’ Christian chart and its songs led Billboard’s Gospel Digital Song Sales in late 2025. The controversy: no testimony or church stood behind the music, artists called it spiritually empty, and a real worship leader shares the name (1)(2)(4).

Is it sinful for a church to sing an AI-generated worship song?
The deeper issue is category, not contamination. Worship is offering, and offering requires a witness, a congregation that owns the words, and real cost to the offerer. A fully generated track has none of these. Churches should apply those three criteria — the Araunah Test — and require disclosure of how songs are made.

Can worship leaders use AI tools at all?
Yes, for scaffolding: chord transcription, rehearsal stems, arrangement drafts a human judges, translation first-drafts refined by human poets, repertoire archives. The line falls at the offering itself — lyrics presented as testimony, vocals presented as a worshiper, or tracks presented as a congregation’s praise must come from actual worshipers (9).

Why did AI worship music chart so easily?
Partly distribution mechanics, but critics across Christian publications agree on a deeper cause: much commercial worship had become formulaic — interchangeable templates written for playlists rather than congregations — and templates are precisely what generative models reproduce best. The machine exposed the genre’s drift toward product (5)(6).

How does AI music affect East Africa’s gospel boom?
Sub-Saharan gospel streaming grew about 50% year-on-year, with Kenya up 30% — a boom cheap generated catalogs will target. But deeply local, multilingual, congregational worship is the hardest music to imitate. The regional strategy: write more indigenous songs, use AI only for scaffolding, and adopt disclosure standards early (7)(10).

Related Reading

• The East African Creative Economy: The Continent’s Next Export
• The Faith Economy Is an Investable Sector
• Sub-Creators with Silicon: A Theology of Building AI
• Should a Pastor Let AI Draft the Sermon?

Sources and Evidence

1. Religion Unplugged — “As AI Music Climbs the Charts, Church Leaders Urge Caution and Discernment” — January 2026 survey of the Solomon Ray controversy, leader responses, the “soul-less music” critique, and the consensus on labeling; Religion Unplugged is a nonprofit religion-journalism outlet.
2. Billboard Pro — “How a MAGA Rapper Used AI to Create a Gospel Song That Climbed the Charts” — Industry-of-record reporting on the creation pipeline, chart performance, and distribution mechanics behind Solomon Ray.
3. Premier Christian News — “AI-made singer tops Christian charts as other artists warn ‘it has no Holy Spirit'” — Source for Forrest Frank’s objection and working artists’ responses.
4. Christianity Today — “The Current No. 1 Christian Artist Has No Soul” — Flagship evangelical magazine’s analysis, including the real Solomon Ray’s “costly worship” response.
5. Baptist Press — “Does AI-generated worship … generate worship?” — Southern Baptist news service’s theological examination, including the Suno training-data and copyright questions.
6. Baptist News Global — “What AI worship exposes about our songs and our souls” — The argument that the controversy reveals how formulaic commercial worship had already become.
7. Afro Soundtrack — “Spotify Wrapped 2025: What the Data Reveals About Gospel Music’s Rise in Sub-Saharan Africa” — Streaming-data analysis: ~50% year-on-year gospel growth in Sub-Saharan Africa, Kenya’s 30% rise, and the listener shift toward local-language artists.
8. TheGrio — “Fair or foul? Creator of AI gospel artist Solomon Ray hits back at critics” — Topher’s defense of the project in his own words.
9. Renewing Worship (Kenny Lamm) — “Is AI-generated worship music acceptable in the church?” — North Carolina Baptist worship strategist’s discernment framework: assist the craft, never replace the heart and witness.
10. The Independent (Uganda) — “Mungu Feni and the new gospel music economy” — Ugandan reporting on the streaming-first gospel career model and rising production investment in the region.