Every procurement tool on the market now claims AI. A chatbot that drafts RFQ emails, a dashboard with a summarize button, a point solution with an LLM stapled to a workflow that hasn't changed in a decade. If you lead sourcing for a manufacturer, you've sat through these demos, and your skepticism is earned.
But the skepticism is aimed at the wrong layer. The problem in direct materials sourcing was never a missing AI feature. The problem is that direct materials never got an operating system.
Indirect got the software. Direct got the spreadsheets.
Over the past twenty years, indirect procurement got its suites: P2P, catalogs, intake, spend analytics. That software works because indirect spend is transactional and repeatable.
Direct materials is neither. A sourcing event for a vehicle platform, a satellite bus, or a server rack starts with an engineering spec, runs through hundreds of line items and dozens of suppliers, and carries constraints that shift with every design revision. The tools that run indirect were never built for that, and ERP only picks up the record after the decisions are made. So the actual work of direct sourcing, the quoting, the supplier back-and-forth, the award tradeoffs, still lives in spreadsheets, inboxes, and institutional memory.
That's the environment teams are now being asked to point AI at. It fails predictably. A model with no access to your items, your suppliers, or your history can only generate plausible text. Ask it about supplier concentration and it approximates. Ask it to build a quote format and it starts from zero. And even when the reasoning is good, it can't act: it cannot ingest your BOM, configure your RFX, or get a supplier to respond. Reasoning without execution is a demo, not a tool.
What an operating system changes
An AI-native direct materials operating system is a different premise. Not AI added to sourcing software, but a single system where engineering, procurement, and suppliers work from the same live data, from first spec through final award, with AI built into every stage of that workflow.
Three things follow from that architecture, and they're the tests worth applying to any vendor claiming AI in this space.
The AI has context. It works from a memory layer built from your operation: your spend, your suppliers, your items, your past projects. When it builds an RFX, it draws on the quote formats your team has actually used, not a generic template.
The AI is grounded. When it answers a question about your data, it fetches the real numbers rather than generating an approximation. Retrieval is separated from transformation, which is what keeps your figures your figures. Any vendor claiming "no hallucinations" without explaining that mechanism is asking you to take it on faith.
The AI executes. It ingests the raw BOM, structures the items, configures the RFX, chases the suppliers, and optimizes the award against your constraints on cost, location, and lead time. The work gets done inside the system where the decision gets made, and every cycle it runs makes the system smarter for the next one.
This isn't theoretical. RFX setup that took two weeks now takes 15 minutes. These are production numbers from real sourcing teams, not pilot metrics.
The window for this decision is now
Every sourcing organization in automotive, electronics, space, and data infrastructure will make an AI architecture decision in the next couple of years, whether deliberately or by accumulation of point tools. The teams that treat it as an operating system decision, one system of record where AI has context, grounding, and the ability to act, will compound an advantage every sourcing cycle. The teams that bolt AI onto fragmentation will get better-worded emails.
See it run, live
On Wednesday, July 29, we're launching LightSource AI Workforce and demoing the full cycle live on real program data: a raw BOM in, a cleaned and structured item list, a complete RFX, activated suppliers, and an optimized award. Idan Mintz, our CTO and co-founder along with Peter DiFalco, Lead Solutions Engineer will walk through how the architecture keeps AI away from your numbers, and then we'll take open questions. Bring the skeptical ones. That's the point of doing it live.
Register for the July 29 launch webinar
9:00 am PT / 12:00 pm ET, 60 minutes. Everyone who registers gets the recording.
Every procurement tool on the market now claims AI. A chatbot that drafts RFQ emails, a dashboard with a summarize button, a point solution with an LLM stapled to a workflow that hasn't changed in a decade. If you lead sourcing for a manufacturer, you've sat through these demos, and your skepticism is earned.
But the skepticism is aimed at the wrong layer. The problem in direct materials sourcing was never a missing AI feature. The problem is that direct materials never got an operating system.
Indirect got the software. Direct got the spreadsheets.
Over the past twenty years, indirect procurement got its suites: P2P, catalogs, intake, spend analytics. That software works because indirect spend is transactional and repeatable.
Direct materials is neither. A sourcing event for a vehicle platform, a satellite bus, or a server rack starts with an engineering spec, runs through hundreds of line items and dozens of suppliers, and carries constraints that shift with every design revision. The tools that run indirect were never built for that, and ERP only picks up the record after the decisions are made. So the actual work of direct sourcing, the quoting, the supplier back-and-forth, the award tradeoffs, still lives in spreadsheets, inboxes, and institutional memory.
That's the environment teams are now being asked to point AI at. It fails predictably. A model with no access to your items, your suppliers, or your history can only generate plausible text. Ask it about supplier concentration and it approximates. Ask it to build a quote format and it starts from zero. And even when the reasoning is good, it can't act: it cannot ingest your BOM, configure your RFX, or get a supplier to respond. Reasoning without execution is a demo, not a tool.
What an operating system changes
An AI-native direct materials operating system is a different premise. Not AI added to sourcing software, but a single system where engineering, procurement, and suppliers work from the same live data, from first spec through final award, with AI built into every stage of that workflow.
Three things follow from that architecture, and they're the tests worth applying to any vendor claiming AI in this space.
The AI has context. It works from a memory layer built from your operation: your spend, your suppliers, your items, your past projects. When it builds an RFX, it draws on the quote formats your team has actually used, not a generic template.
The AI is grounded. When it answers a question about your data, it fetches the real numbers rather than generating an approximation. Retrieval is separated from transformation, which is what keeps your figures your figures. Any vendor claiming "no hallucinations" without explaining that mechanism is asking you to take it on faith.
The AI executes. It ingests the raw BOM, structures the items, configures the RFX, chases the suppliers, and optimizes the award against your constraints on cost, location, and lead time. The work gets done inside the system where the decision gets made, and every cycle it runs makes the system smarter for the next one.
This isn't theoretical. RFX setup that took two weeks now takes 15 minutes. These are production numbers from real sourcing teams, not pilot metrics.
The window for this decision is now
Every sourcing organization in automotive, electronics, space, and data infrastructure will make an AI architecture decision in the next couple of years, whether deliberately or by accumulation of point tools. The teams that treat it as an operating system decision, one system of record where AI has context, grounding, and the ability to act, will compound an advantage every sourcing cycle. The teams that bolt AI onto fragmentation will get better-worded emails.
See it run, live
On Wednesday, July 29, we're launching LightSource AI Workforce and demoing the full cycle live on real program data: a raw BOM in, a cleaned and structured item list, a complete RFX, activated suppliers, and an optimized award. Idan Mintz, our CTO and co-founder along with Peter DiFalco, Lead Solutions Engineer will walk through how the architecture keeps AI away from your numbers, and then we'll take open questions. Bring the skeptical ones. That's the point of doing it live.
Register for the July 29 launch webinar
9:00 am PT / 12:00 pm ET, 60 minutes. Everyone who registers gets the recording.
Every procurement tool on the market now claims AI. A chatbot that drafts RFQ emails, a dashboard with a summarize button, a point solution with an LLM stapled to a workflow that hasn't changed in a decade. If you lead sourcing for a manufacturer, you've sat through these demos, and your skepticism is earned.
But the skepticism is aimed at the wrong layer. The problem in direct materials sourcing was never a missing AI feature. The problem is that direct materials never got an operating system.
Indirect got the software. Direct got the spreadsheets.
Over the past twenty years, indirect procurement got its suites: P2P, catalogs, intake, spend analytics. That software works because indirect spend is transactional and repeatable.
Direct materials is neither. A sourcing event for a vehicle platform, a satellite bus, or a server rack starts with an engineering spec, runs through hundreds of line items and dozens of suppliers, and carries constraints that shift with every design revision. The tools that run indirect were never built for that, and ERP only picks up the record after the decisions are made. So the actual work of direct sourcing, the quoting, the supplier back-and-forth, the award tradeoffs, still lives in spreadsheets, inboxes, and institutional memory.
That's the environment teams are now being asked to point AI at. It fails predictably. A model with no access to your items, your suppliers, or your history can only generate plausible text. Ask it about supplier concentration and it approximates. Ask it to build a quote format and it starts from zero. And even when the reasoning is good, it can't act: it cannot ingest your BOM, configure your RFX, or get a supplier to respond. Reasoning without execution is a demo, not a tool.
What an operating system changes
An AI-native direct materials operating system is a different premise. Not AI added to sourcing software, but a single system where engineering, procurement, and suppliers work from the same live data, from first spec through final award, with AI built into every stage of that workflow.
Three things follow from that architecture, and they're the tests worth applying to any vendor claiming AI in this space.
The AI has context. It works from a memory layer built from your operation: your spend, your suppliers, your items, your past projects. When it builds an RFX, it draws on the quote formats your team has actually used, not a generic template.
The AI is grounded. When it answers a question about your data, it fetches the real numbers rather than generating an approximation. Retrieval is separated from transformation, which is what keeps your figures your figures. Any vendor claiming "no hallucinations" without explaining that mechanism is asking you to take it on faith.
The AI executes. It ingests the raw BOM, structures the items, configures the RFX, chases the suppliers, and optimizes the award against your constraints on cost, location, and lead time. The work gets done inside the system where the decision gets made, and every cycle it runs makes the system smarter for the next one.
This isn't theoretical. RFX setup that took two weeks now takes 15 minutes. These are production numbers from real sourcing teams, not pilot metrics.
The window for this decision is now
Every sourcing organization in automotive, electronics, space, and data infrastructure will make an AI architecture decision in the next couple of years, whether deliberately or by accumulation of point tools. The teams that treat it as an operating system decision, one system of record where AI has context, grounding, and the ability to act, will compound an advantage every sourcing cycle. The teams that bolt AI onto fragmentation will get better-worded emails.
See it run, live
On Wednesday, July 29, we're launching LightSource AI Workforce and demoing the full cycle live on real program data: a raw BOM in, a cleaned and structured item list, a complete RFX, activated suppliers, and an optimized award. Idan Mintz, our CTO and co-founder along with Peter DiFalco, Lead Solutions Engineer will walk through how the architecture keeps AI away from your numbers, and then we'll take open questions. Bring the skeptical ones. That's the point of doing it live.
Register for the July 29 launch webinar
9:00 am PT / 12:00 pm ET, 60 minutes. Everyone who registers gets the recording.
Faster sourcing. Lower cost. Less chaos.
See how LightSource connects engineering, procurement, and suppliers in one operating system to help you launch faster at lower cost.
SOC 2
Kearney #1 2024
Gartner Cool Vendor
Procuretech 100
G2 Top Rated
Faster sourcing. Lower cost. Less chaos.
See how LightSource connects engineering, procurement, and suppliers in one operating system to help you launch faster at lower cost.
SOC 2
Kearney #1 2024
Gartner Cool Vendor
Procuretech 100
G2 Top Rated
Faster sourcing. Lower cost. Less chaos.
See how LightSource connects engineering, procurement, and suppliers in one operating system to help you launch faster at lower cost.
SOC 2
Kearney #1 2024
Gartner Cool Vendor
Procuretech 100
G2 Top Rated
Trusted by:
Trusted by:
Trusted by:
*GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally, and COOL VENDORS is a registered trademark of Gartner, Inc. and/or its affiliates and are used herein with permission. All rights reserved. Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner’s research organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.