Better Health for All
Not Applicable
Value not applicable to this business. xAI is a software and high-performance computing company focused on generative AI and LLMs; its core business does not involve the development of medical treatments, healthcare services, or public health initiatives.
Fair Money & Economic Opportunity
Not Applicable
Value not applicable to this business. xAI is a software and AI infrastructure company; it does not provide financial services, lending, or wealth management products, making the Fair Money & Economic Opportunity value inapplicable.
Fair Pay & Worker Respect
-10
As an employer, xAI is fully in scope for Fair Pay & Worker Respect; however, its core business of AI development and infrastructure does not inherently advance or harm these labor practices, which are determined by internal management policies. Based on the provided evidence, xAI (XAI.P) is scored on a limited set of KPIs due to its status as a relatively new private entity with restricted public data. For 'labor_violation_incidents', the company receives a tier of 0 as there is no public record of substantiated labor-law breaches or human-rights violations in the provided articles.
1
2
While community organizers in Memphis have expressed skepticism regarding job guarantees and transparency, no legal findings or fines are documented.
3
For 'collective_bargaining_share', the company is tiered at -100.
4
Evidence from highlights that local community organizers and labor advocates characterize the project's development as lacking public accountability, noting that city officials signed NDAs with xAI.
5
Furthermore, the article draws parallels to historical labor suppression in the Memphis region and notes that there are 'no guarantees' for local hiring or labor protections.
6
In the broader context of the AI industry described in , while some tech workers (e.g., Alphabet Workers Union) have successfully organized, there is no evidence of union recognition or collective bargaining at xAI.
7
Given the documented use of NDAs to bypass public scrutiny and the lack of any recognized union presence, the lowest tier is applied.
8
Other KPIs such as 'living_wage_coverage', 'ceo_median_pay_ratio', and 'safety_incident_rate' were omitted as the articles provided only theoretical frameworks for AI in HR or general industry trends (Articles
9
,
10
,
11
,
12
) rather than specific quantitative data for xAI.
Fair Trade & Ethical Sourcing
-10
As a developer of high-performance computing infrastructure and AI hardware, xAI relies on complex global supply chains for semiconductors and GPUs, which are high-risk sectors for forced labor and unethical sourcing of raw materials. While the core business is software-focused, the physical infrastructure requirements necessitate active management of supply chain ethics. xAI (XAI.P) is evaluated based on its operational activities and the broader supply chain risks associated with its hardware infrastructure (data centers and supercomputers). Regarding forced and child labor, while the company operates in a high-risk sector for hardware components (specifically minerals like cobalt), there are no substantiated reports of labor violations directly linked to xAI's specific supply chain in the provided evidence.
1
The company is noted to be under pressure to align with ethical labor practices, and its founder has publicly proposed transparency measures (e.g., webcams in mines) to prevent child labor, suggesting a strict enforcement stance (Tier -10).
2
In terms of materials risk, xAI's infrastructure relies on high-performance computing hardware.
3
Evidence indicates that major tech companies in this ecosystem (including those linked to xAI's leadership) acknowledge that smelters in their supply chains process minerals from conflict-affected regions like the DRC.
4
Specifically, the xAI facility in Memphis and the Colossus supercomputer utilize tens of thousands of GPUs, which are intensive in high-risk minerals.
5
While the company is subject to active controls and industry-wide pressure to mitigate these risks, the high-risk spend remains significant (Tier -60).
6
Ethical clause coverage is scored at 0 (N/A) because the provided articles do not contain specific data on xAI's internal procurement contracts or the percentage of suppliers bound by enforceable ethical clauses.
7
8
While general regulatory pressures (EU AI Act, GDPR) are mentioned, they do not constitute evidence of specific contractual clauses for xAI.
9
10
Other KPIs such as audit frequency, traceability, and remediation speed were omitted due to a lack of company-specific evidence in the provided text.
11
12
Honest & Fair Business
-20
As a software and AI infrastructure company, xAI's core business does not inherently advance or harm honest business practices; its score will depend on its future transparency, data handling, and regulatory compliance regarding AI safety and disclosure. xAI (XAI.P) demonstrates a complex profile regarding 'Honest & Fair Business' based on the provided evidence. Regarding regulatory compliance, while no specific dollar-value fines are confirmed, the company has received significant warning signals and is under active investigation by the European Union (Digital Services Act), France, and the UK for illegal content and risk management failures.
1
It has also been removed from markets in Malaysia and Indonesia due to safety concerns.
2
3
Consequently, it is tiered at -10 for regulatory_fines_3y to reflect these formal warnings and investigations. On transparency, the company scores extremely poorly. Independent analysis from Stanford (CRFM) assigned the company a score of 0 for most transparency indicators.
4
Furthermore, xAI is actively litigating against transparency laws in California (AB 2013) and Colorado, arguing that disclosing training data sources would destroy trade secrets.
5
This resistance to independent disclosure mandates justifies a -100 tier for transparency_index. For whistleblower_policy, the company claims to have a policy allowing anonymous reporting with anti-retaliation protections.
6
7
However, external assessments note a lack of detail on investigation processes or independent oversight, and the platform has reportedly 'hollowed out' its trust and safety staffing.
8
9
It is tiered at -10 as a comprehensive policy is communicated but lacks evidence of robust independent enforcement. Third-party verification is tiered at -70. While xAI mentions working with entities like NIST and RAND, it reserves the right to redact findings and has been criticized by civil society organizations for protections that fall below industry norms.
10
11
12
Independent assessments by Common Sense Media found significant safety failures (e.g., Grok generating millions of nonconsensual sexualized images), contradicting the company's internal safety claims.
13
Kind to Animals
Not Applicable
Value not applicable to this business. xAI is a software and high-performance computing company; its core business activities do not involve animal products, animal testing, or direct interaction with animal welfare.
No War, No Weapons
-30
As a developer of general-purpose large language models, xAI's core business is not inherently tied to the defense or weapons industry, though its high-performance computing infrastructure could theoretically be dual-use in the future. Based on the provided evidence from 2026, xAI has actively positioned itself as a primary military AI provider by removing ethical safeguards that restrict lethal applications. Regarding **dual_use_technology** and **ai_military_safeguards**, xAI signed an agreement with the U.S. Department of Defense (DOD) to integrate its Grok models into classified military systems for "all lawful purposes."
1
This specifically includes weapons development, battlefield operations, and potentially powering fully autonomous military drones.
2
This stands in direct contrast to competitors like Anthropic, which maintained safeguards against such uses.
3
Consequently, xAI is scored at -100 for these KPIs as it has deployed AI for lethal applications without technical safeguards and lacks an external ethics board prohibiting military use. For **ethical_red_lines_coded**, the company is scored at -100 because it explicitly allows "full usage of its models to be utilized with defense systems," rejecting the restrictive ethical frameworks (red lines) adopted by other AI firms regarding autonomous armed drones and mass surveillance.
4
In terms of **surveillance_transparency**, the company receives a -100 as it has entered into classified contracts for sensitive intelligence work without public disclosure of the specific scopes or transparency measures, unlike humanitarian-only models.
5
Finally, for **revenue_arms_contracts**, the company is tiered at -30. While the articles confirm signed contracts with the DOD and the U.S. General Services Administration—identifying these as a financial "lifeline"—there is no specific percentage of total revenue provided.
6
Given the high burn rate ($1 billion/month) and the recent nature of the December partnership, the evidence supports the existence of defense contracts but does not yet confirm they exceed 5% of total revenue.
7
Planet-Friendly Business
-50
The company's core business involves operating massive, energy-intensive high-performance computing clusters (such as the Colossus supercomputer), which inherently creates a significant carbon footprint and high electricity demand, posing a challenge to environmental stewardship. xAI’s environmental performance is characterized by significant fossil fuel reliance and ongoing regulatory non-compliance. Regarding emissions, the company’s Southaven site permit application estimates greenhouse gas emissions exceeding 6 million tons per year, placing it in the -80 tier for high-intensity footprints with limited disclosure.
1
There is no evidence of SBTI-aligned targets; instead, reports indicate the company is bypassing standard environmental reviews, warranting a -90.
2
3
In terms of energy, xAI relies heavily on methane-powered gas turbines.
4
At the Memphis site, it operates 33 turbines, and at Southaven, it has installed 27.
5
6
While the industry average for new AI data centers is 75% natural gas, xAI’s current operational setup described in the articles shows 0% renewable energy integration for these primary power sources, resulting in a -100.
7
8
Environmental compliance is a major area of concern. Multiple sources (EPA, SELC, and local health departments) confirm that xAI has operated dozens of gas turbines without required permits or pollution controls in both Memphis and Southaven.
9
10
11
12
13
This pattern of recurrent unpermitted operations and the resulting regulatory scrutiny justifies a -100 for compliance violations.
14
15
Finally, regarding climate justice, the company’s operations are located in South Memphis and Southaven—areas with predominantly Black populations and existing high rates of asthma and industrial pollution.
16
17
18
Despite these impacts and local petitions for shutdowns, there is no evidence of formal climate justice programs or community co-design, leading to a -90.
19
20
While xAI has proposed an $80 million greywater facility to reduce aquifer strain by 9%, this is a future project (2026–2027) and does not yet offset current operational impacts.
21
Respect for Cultures & Communities
-20
As a developer of high-performance computing infrastructure (Colossus supercomputer) and AI models, xAI relies on a complex hardware supply chain involving minerals like cobalt and lithium, which are frequently linked to community displacement and human rights risks in extraction regions. While the core business is software-focused, the physical infrastructure requirements place it in scope for potential supply-chain impacts on communities. xAI’s operations, specifically the Colossus supercomputer in Memphis, TN, demonstrate significant negative impacts on the local majority-Black community of Boxtown/South Memphis. **Community Grievance Resolution & Transparency:** The company is scored -100 due to a documented lack of a functional grievance mechanism and active avoidance of community engagement.
1
2
Reports indicate xAI used NDAs with city officials to bypass public input, ignored requests for public meetings, and allegedly operated unpermitted turbines in violation of the Clean Air Act.
3
4
5
**Supply Chain & Community Harm:** Scored -80 based on documented environmental justice harms.
6
The facility operates dozens of methane gas turbines in an area already designated an "asthma capital" with cancer risks 4x the national average.
7
8
9
These turbines emit nitrogen oxides and formaldehyde, worsening air quality in a historically marginalized community.
10
11
12
**Water Rights Conflicts:** Scored -50. The facility draws 1–1.5 million gallons of water daily from the Memphis Sand Aquifer (projected to rise to 5+ million).
13
14
15
Experts warn this over-extraction risks pulling arsenic and other contaminants into the city's drinking water.
16
17
18
While xAI committed to an $80M greywater plant, this is a partial remediation for a conflict it created.
19
20
21
**Community Protest Severity:** Scored -60. Multiple organizations (NAACP, MCAP, SELC) have organized sustained opposition, including legal appeals to halt operations and raucous public hearings where company representatives were booed.
22
23
24
**Economic Factors:** Scored -30 for local employment as activists and experts note data centers require minimal staffing and there are no binding local hire commitments.
25
26
27
Revenue sharing is scored -20; while the Mayor proposed a 25% tax allocation to the local zip code, this is a city-led initiative rather than a verified company-governed benefit-sharing agreement.
28
29
Safe & Smart Tech
-10
As a developer of generative AI and LLMs, xAI's core business is directly subject to the risks and responsibilities of Safe & Smart Tech, including algorithmic bias, data privacy, and AI safety, but its inherent impact depends on its specific deployment and governance practices rather than the technology itself. xAI exhibits a polarized profile in Safe & Smart Tech, with strong technical infrastructure and data retention policies offset by significant security incidents and governance failures. On the negative side, the company suffered a significant data exposure incident in August 2025, where over 370,000 Grok conversations (including passwords, medical details, and malware code) were indexed by search engines due to a lack of disclosure regarding the 'share' function.
1
This reflects poorly on vulnerability management and user data control, as warnings from users had persisted since January 2025 without remediation.
2
Furthermore, xAI's 'default-on' training policy for X user data, which lacks an opt-out in mobile apps, has drawn criticism for bypassing explicit consent.
3
4
Governance is a major weakness; federal assessments (GSA) concluded Grok-4 failed safety and alignment expectations, posing 'elevated safety risks.'
5
6
Independent audits by Holistic AI found a 97.3% failure rate in resisting jailbreaking prompts, significantly lagging behind peers like OpenAI.
7
On the positive side, xAI demonstrates strong data minimization for enterprise users, with a standard 30-day deletion policy for inputs/outputs and a commitment not to use business data for training.
8
9
10
The company utilizes advanced infrastructure (Colossus supercluster) and maintains a baseline security framework (NIST 800-171 Rev.3).
11
12
13
It also provides some transparency by open-sourcing model weights (Grok-1) and system prompts.
14
15
While it maintains a bug bounty program and conducts annual third-party penetration testing, these measures failed to prevent the massive indexing exposure, capping its security scores.
16
17
18
Zero Waste & Sustainable Products
Not Applicable
Value not applicable to this business. As a pure-software and high-performance computing infrastructure firm, xAI does not produce physical consumer goods, packaging, or products subject to cradle-to-cradle waste management cycles.