Ethical Stack.

IBM watsonx.ai

Enterprise model platform for hosted foundation-model inference, prompt workflows and tuning.

Score calculation

(3 + 8 + 4 + 6 + 3) ÷ 5 = 4.8

Pressure target: −1 overall point. Overall score: 3.8/10.

5 of six dimensions rated. Unrated dimensions are excluded from the average.

Full scoring method ↗
Setup details

watsonx.ai SaaS foundation-model API and Prompt Lab; excludes self-managed watsonx deployments and third-party cloud services outside IBM’s described route.

What fits your needs

watsonx.ai SaaS foundation-model API and Prompt Lab; excludes self-managed watsonx deployments and third-party cloud services outside IBM’s described route.

Where it fits

Use the SaaS API or Prompt Lab to test hosted foundation models in a regional IBM account.

See the evidence
Tradeoffs to expect

Route and limits

You choose the model and region; optional saved assets and governance monitoring have distinct storage behavior.

See the evidence
How to switch

Try it

Create a watsonx.ai project, select a regional endpoint and model, then test the API or Prompt Lab.

See the evidence
Practical questions

Before relying on it

Review account region, selected model provider, asset-saving choices, governance monitoring and project access.

See the evidence

Energy practices

3/10↑

Public docs establish the IBM watsonx.ai SaaS inference endpoints and customer-selected region route; they don’t show its request-level physical footprint. The API lets customers choose a model ID and request generation parameters; model selection and token-based usage are practical controls, without measured energy savings.

See the evidence

This practices and disclosure grade covers IBM watsonx.ai SaaS inference endpoints and customer-selected region. Public product/API documentation establishes how this route operates and the listed user controls, but gives no measured per-request energy, emissions or water. A corporate electricity claim is not allocated to this route.

1 + 1 reporting + 0 electricity + 1 controls = 3/10. This grades documented practices and transparency.

Request-level electricity, emissions, water use, serving utilization and any route-specific clean-power matching are not reported for IBM watsonx.ai SaaS inference endpoints and customer-selected region.

Reporting1/3
Official documentation describes the IBM watsonx.ai SaaS inference endpoints and customer-selected region operating boundary, but does not publish a dated quantitative energy, emissions or water result for this route.
Electricity sourcing0/3
The reviewed sources do not map IBM watsonx.ai SaaS inference endpoints and customer-selected region to a site-specific electricity mix or time-matched clean-power supply.
Resource controls1/3
The API lets customers choose a model ID and request generation parameters; model selection and token-based usage are practical controls, without measured energy savings.

Water disclosure

Water reporting has not been assessed for this choice.

Training data

—Depends on the model↑

watsonx.ai routes requests to customer-selected foundation models; training data provenance and usage rights vary by model publisher.

See the evidence

IBM’s security page describes handling of submitted prompts and outputs but doesn’t confirm the training data or permissions for the customer-selected foundation models. Prompt privacy and original model training data are separate questions.

Selected foundation model, training data inventory, source permissions, tuning data rights and publisher-specific terms need model-level review.

Your data & privacy

8/10↑

IBM says foundation-model prompts and outputs are not logged and are not used to improve IBM models; assets are stored only when users choose to save them. It notes IBM Cloud or AWS hosting and optional governance monitoring.

See the evidence

IBM’s watsonx.ai security documentation says prompts and outputs submitted through Prompt Lab or API are not logged and are not used to improve IBM models; unsaved prompt text remains in browser cache, while saved assets are stored in the customer project bucket. Optional watsonx.governance monitoring can store payloads in a customer-administered database.

Clear prompt logging and training-use statements plus user-controlled saving, with documented exceptions for optional governance monitoring and deployment route.

The chosen model, regional endpoint, saved asset configuration, governance monitoring, and any model-specific terms should be checked for the deployed account.

Openness

4/10↑

API interfaces and model selection are documented, with managed service code and infrastructure remaining controlled by IBM and its cloud partners.

See the evidence

watsonx.ai exposes REST APIs and model catalogs, but the managed SaaS serving runtime is not supplied as a self-hostable API. IBM has separate software deployments and open-source model access that are outside this SaaS scope.

Documented models and APIs with provider-controlled SaaS runtime.

Self-managed IBM software and selected third-party foundation models have separate openness terms.

Worker protections

6/10↑

IBM publishes global employee standards, workforce indicators, employee reporting channels, and supplier conduct expectations; none is specific to Granite training labor.

See the evidence

IBM reports pay-equity and workforce-safety indicators and sets standards for wages, hours, freedom of association, and suppliers. It also describes employee and supplier reporting channels with non-retaliation. These group-level protections don’t show conditions for Granite data, evaluation, or infrastructure workers. These findings concern the developer or operator. This product’s use in a particular deployment needs separate evidence.

1 base + 2 conditions + 2 voice and remedy + 1 assurance = 6/10. IBM documents group-level systems; evidence for Granite labor and independent outcomes is limited.

The sources do not confirm which Granite contributors or suppliers are covered, or publish AI-workforce-specific pay, locations, grievance outcomes, or independent labor-audit results.

Pay & conditions2/3
IBM publishes global employee standards on wages, hours, freedom of association, safety and supplier expectations, plus aggregate workforce metrics; it does not show model-specific coverage.
Worker voice & remedy2/3
IBM describes Open Door/Speak-Up and external channels, review of reports, and non-retaliation; no Granite-specific grievance data is disclosed.
Evidence & follow-through1/3
IBM describes compliance reviews and supplier standards, but these sources do not provide independent worker-level assurance or remediation outcomes for Granite labor.
Workers rubric ↗

Military ties

3/10Broad defense ties↑

IBM publicly markets defense AI and was selected for the Missile Defense Agency SHIELD contract; neither source establishes Granite deployment.

See the evidence

IBM says it will support the MDA SHIELD program and promotes AI for defense and intelligence work. Granite weights use Apache 2.0 terms without a military-use prohibition in the reviewed card; no source links this checkpoint to that contract or describes its military end-use oversight. These findings concern the developer or operator. This product’s use in a particular deployment needs separate evidence.

1 + 0 use restrictions + 0 enforcement & oversight + 2 relationship disclosure = 3/10. Known defense ties lower this rating; safeguards can improve it within the relationship’s cap.

No Granite-specific defense contract, deployment, military-use restriction, or downstream enforcement record was established.

Use restrictions0/4
The Granite card lists Apache 2.0 and no model-specific military-use restriction; the license is not a downstream deployment control.
Enforcement & oversight0/3
IBM describes governance for defense customers, but the sources don’t show monitoring or enforcement over use of downloadable Granite weights.
Relationship disclosure2/2
IBM publicly identifies its SHIELD selection and markets AI-enabled defense and intelligence solutions.

IBM publicly supports the MDA SHIELD program and sells AI-enabled defense/intelligence services. This is a company relationship; no Granite checkpoint deployment is established. The relationship limits this grade to 3/10.

Military rubric ↗

BDS targets

Pressure target↑

IBM is named in the BNC No Tech campaign’s discussion of technology companies; this is a scoped pressure-campaign match, not a universal BDS status.

See the evidence

The official BNC No Tech campaign page names IBM alongside other technology companies in its discussion of systems serving the Israeli military. This supports a pressure-campaign match at the IBM parent-company level; it doesn’t show a use or contract involving every Granite checkpoint. These findings concern the developer or operator. This product’s use in a particular deployment needs separate evidence.

This match is limited to the named companies in the BNC No Tech campaign. It does not cover every BDS campaign, local group, investment, supplier, or product deployment; lists and ownership can change.

More from IBM

All sources

Side by side