Gotham

Palantir Gotham is an AI-ready, closed-source operating system for national security, designed to improve and accelerate decisions for operators across all echelons and domains — land, sea, air, space, and cyberspace.

Reviewed by 7wData

On this page

Publisher review

Palantir Gotham is an AI-ready, closed-source operating system for national security, designed to improve and accelerate decisions for operators across all echelons and domains — land, sea, air, space, and cyberspace. It is purpose-built for global defense agencies, intelligence groups, disaster relief organizations, and law enforcement entities, including the U.S. Department of Defense, the FBI, and allied military commands.

Gotham transforms raw, complex data into a comprehensive, up-to-date understanding of the world, enabling users to task sensors (from drones to satellites), pair targets with effectors, and manage kill chains with human-in-the-loop or AI-driven rules. The platform integrates massive volumes of near-real-time data and presents them in a unified interface, turning any bunker or outpost into an instant command center with mixed-reality capabilities. Pricing starts with a nominal £1.00 trial, then scales through pilot licences (£50,000–£500,000), organisation licences (£3,000,000–£2,500,000), and usage-based tiers (£5,000–£2,500,000), plus support services from £12,100 per person per month (FSR-1) to £38,500 (FSR-4).

Perpetual licences cost £147,324, appliance options £157,800, and monthly cloud licences £7,366.20. Gotham is not a general-purpose analytics tool; it is a defense-specific decision-support system that demands deep integration and carries significant lock-in risk, as noted by critics who highlight its lack of substitutability, inspectability, and interoperability. Competitors include DataWalk, Zoho Analytics, dbt, and Teradata VantageCloud, though none match Gotham's military-grade geospatial and targeting workflows.

The platform's strengths — real-time sensor tasking, AI-powered kill chain automation, mixed-reality command centers, and proven use by elite defense units — are offset by its extreme cost, vendor dependency, closed-source architecture, and limited commercial adoption outside government contracts. For organizations that can afford the price and tolerate the lock-in, Gotham provides a unique operational advantage in contested environments.

Get the AI & data signal, daily.

48k+ subscribers read this every morning. One email, both newsletters. Unsubscribe anytime.

How it works

  1. AI-powered kill chain

    Seamlessly integrates target identification and effector pairing, supporting soldiers with an AI-driven kill chain for responsible target management.

  2. Autonomous sensor tasking

    Tasks drones, satellites, and other sensors based on AI rules or manual inputs, maximizing asset effectiveness in dynamic operational environments.

  3. Unified data integration

    Ingests massive volumes of near-real-time data from disparate sources and presents them in a single, comprehensive interface for operators.

  4. Mixed-reality command center

    Transforms any bunker or outpost into an instant command center with mixed-reality capabilities, bridging physical and digital worlds.

  5. Cross-domain decision support

    Supports operators at every echelon and domain — land, sea, air, space, and cyberspace — with tailored interfaces for each mission context.

  6. Human-in-the-loop control

    Enables manual oversight of AI-driven processes, ensuring responsible use of autonomous systems while maintaining operator authority.

  7. Scalable pilot and organisation licences

    Offers tiered pilot licences from £50,000 to £500,000 and organisation licences from £3,000,000, with usage-based add-ons up to £2,500,000.

Strengths and trade-offs

Strengths

  • Gotham is deployed by the U.S. Department of Defense, FBI, and allied intelligence agencies for real-world combat and intelligence operations.
  • The platform enables autonomous sensor tasking of drones and satellites using AI-driven rules, reducing manual workload in dynamic battlespaces.
  • Mixed-reality capabilities allow operators to turn any bunker or outpost into an instant command center, enhancing situational awareness.
  • Pricing starts at a nominal £1.00 trial, with pilot licences from £50,000, making initial evaluation accessible for qualified government customers.

Trade-offs

  • Gotham is extremely expensive, with organisation licences starting at £3,000,000 per year and usage-based fees reaching £2,500,000.
  • The platform is closed-source, creating vendor lock-in that makes it difficult and costly to migrate data or workflows to alternative systems.
  • Commercial adoption is very limited; Gotham's primary market is government and defense, not private-sector enterprises.
  • Support services add significant ongoing costs, with field service representatives priced from £12,100 to £38,500 per person per month.

Pricing context

Trial: £1.00. Pilot licences: £50,000–£500,000. Organisation licences: £3,000,000 (PT-ORG) and £2,500,000 (PT-ORG-ADD).

Usage tiers: £5,000–£2,500,000. Support services: IMPL-REP £150,000/quarter, FSR-1 £12,100/month, FSR-2 £18,700/month, FSR-3 £29,700/month, FSR-4 £38,500/month. Perpetual licence: £147,324.

Appliance: £157,800. Monthly cloud licence: £7,366.20. Cloud support: £21,500/year.

Getting started with Gotham

  1. Sign up for trial

    Visit the Palantir Gotham website and complete the registration form for a £1.00 trial. Provide your government or defense agency credentials to verify eligibility. Await approval and receive login credentials for the trial environment.

  2. Connect data sources

    In the Gotham interface, navigate to the data integration panel. Add connectors for your intelligence feeds, sensor streams, and databases. Configure authentication and ingestion schedules to pull near-real-time data into the unified platform.

  3. Configure targeting rules

    Set up AI-driven kill chain rules by defining target criteria and effector pairings. Use the rule builder to specify conditions for autonomous sensor tasking. Enable human-in-the-loop oversight for critical decisions.

  4. Run a sensor tasking mission

    Select a drone or satellite asset from the sensor inventory. Assign it to a target area using the tasking interface. Monitor the mission in real time through the mixed-reality command center, adjusting parameters as needed.

  5. Deploy to operational use

    After successful trial, contact sales to purchase a pilot licence starting at £50,000. Work with Palantir field service representatives to integrate Gotham into your command infrastructure. Schedule ongoing support at the appropriate FSR tier.

Frequently Asked Questions

What is Palantir Gotham and who uses it?

Palantir Gotham is an AI-ready, closed-source operating system for national security used by the U.S. Department of Defense, FBI, and allied intelligence agencies. It helps operators make faster decisions across land, sea, air, space, and cyberspace by transforming raw data into actionable intelligence.

How does Palantir Gotham's AI kill chain work?

Gotham's AI-powered kill chain integrates target identification with effector pairing, supporting soldiers in responsible target management. It uses AI-driven rules or manual inputs to automate the process, enabling operators to manage kill chains efficiently while maintaining human-in-the-loop control for oversight.

What is the pricing for Palantir Gotham?

Gotham pricing starts with a £1.00 trial, then pilot licences from £50,000 to £500,000, and organisation licences from £3,000,000 per year. Usage-based tiers range from £5,000 to £2,500,000. Support services cost £12,100 to £38,500 per person per month, with perpetual licences at £147,324.

What are the risks of vendor lock-in with Gotham?

Gotham is closed-source, creating significant vendor lock-in. Critics highlight its lack of substitutability, inspectability, and interoperability, making it difficult and costly to migrate data or workflows to alternative systems. This lock-in risk is a major consideration for organizations evaluating the platform.

Can Gotham task drones and satellites autonomously?

Yes, Gotham enables autonomous sensor tasking of drones, satellites, and other sensors based on AI rules or manual inputs. This maximizes asset effectiveness in dynamic operational environments, reducing manual workload for operators in contested battlespaces.

What are the main competitors to Palantir Gotham?

Competitors include DataWalk, Zoho Analytics, dbt, and Teradata VantageCloud. However, none match Gotham's military-grade geospatial and targeting workflows. Gotham's primary market is government and defense, with limited commercial adoption outside government contracts.

Alternatives

How Gotham compares

Direct head-to-head against 3 competitors. Picked by 7wData.

This tool

Gotham

Pricing
Trial: £1.00. Pilot licences: £50,000–£500,000. Organisation licences: £3,000,000 (PT-ORG) and £2,500,000 (PT-ORG-ADD). Usage tiers: £5,000–£2,500,000. Support services: IMPL-REP £150,000/quarter, FSR-1 £12,100/month, FSR-2 £18,700/month, FSR-3 £29,700/month, FSR-4 £38,500/month. Perpetual licence: £147,324. Appliance: £157,800. Monthly cloud licence: £7,366.20. Cloud support: £21,500/year.
Target
Palantir Gotham is an AI-ready, closed-source operating system for national security, designed to improve and accelerate decisions for operators across all echelons and domains —
Strength
Gotham is deployed by the U.S. Department of Defense, FBI, and allied intelligence agencies for real-world combat and intelligence operations.
Watch for
Gotham is extremely expensive, with organisation licences starting at £3,000,000 per year and usage-based fees reaching £2,500,000.

Metropolis

Pricing
Free
Target
Designers seeking Gotham-like fonts
Deployment
Download
Strength
Closest free alternative to Gotham
Watch for
Limited stylistic variations

Figtree

Pricing
Free
Target
Web designers
Deployment
Google Fonts
Strength
Variable font with 7 weights
Watch for
Less stocky than Gotham

Montserrat

Pricing
Free
Target
General-purpose design
Deployment
Google Fonts
Strength
9 weights + italics
Watch for
Overused in design

User reviews

No user reviews yet. Be the first to write one.

Sources

Reporting on this tool draws on these publicly available sources.

  1. www.palantir.com
  2. assets.applytosupply.digitalmarketplace.service.gov.uk
  3. hash.ai
  4. financhle.com
  5. www.reddit.com
  6. www.reddit.com