Meta
Meta AI / Llama
Open-weight frontier you can self-host.
Impact Indicators
Job Displacement
No verified job-loss data exists for Meta AI or Llama. The available peer-reviewed evidence measures broader generative-AI effects on freelance labor demand, not displacement caused by this model. Meta's own layoffs and restructuring are not evidence that Meta AI or Llama displaced work.
Privacy & Surveillance
Meta AI has drawn regulator scrutiny in Ireland over Meta's use of public adult Facebook and Instagram content to train its AI; the regulator's own statement describes engagement and continued scrutiny, not a finding, enforcement action or violation. Separately, personal prompts shared through Meta AI's Discover feed were publicly visible, and Meta patched a flaw that could have exposed other users' prompts and responses, with no confirmed exploitation reported.
Water Usage
No measured water figure exists for this model, for Llama generally, or for the Meta AI product. Meta reports company and data-center water figures, but those totals cannot be attributed to this model, withdrawal is not the same as consumption, and restoration volumes do not reduce consumption. The remaining industry evidence is modeled and estimated, not measured for this model.
Energy & Carbon
No measured energy or carbon figure exists for this model. Meta's published training figures describe an earlier Llama version and are retained only as research on a different model version; they are not transferred here. Meta's company-level energy and market-based emissions reporting is an accounting result for its overall operations, not this model's measured load, and the remaining industry evidence is modeled.
U.S. reporting requirements do not yet provide standardized, model-level disclosure for many AI human and environmental impacts. When U.S. evidence is incomplete, DoctorKnow uses verifiable EU regulatory disclosures, public-sector data, independent research, and reputable journalism to provide the strongest evidence currently available.
Evidence scope is identified as model-specific, provider-level, facility-level, regional/grid, or industry-level.
Scores use a 0–100 scale. Higher scores indicate stronger performance or lower concern; 100 is the best possible rating.
Reasoning
89
Coding
88
Writing
87
Analysis
87
Speed
74
Scores use a 0–100 scale, where 100 represents the strongest outcome and 0 the weakest.
What Meta AI / Llama does
Meta AI / Llama is built by Meta and is positioned for code, writing, analysis and agents. Our readings score it strongest on multi-step reasoning and code generation.
Its weakest measured axis is response speed (74/100), so plan around that when the work depends on it. Access is open, and the weights are published, so the model can be inspected and self-hosted.
Ethical stance
Transparency reads 79/100. Documentation, evaluations and model details are unusually open for the category.
Labor and job-displacement posture reads 38/100. Data-annotation conditions and displacement impact are largely unaddressed.
Regulatory posture reads 44/100. The vendor cooperates with regulators selectively, and lobbies against parts of it.
Privacy and data handling
Privacy and data handling reads 72/100. Retention is short or optional, training on your prompts is off by default, and enterprise terms are clear.
Because the weights are published, the strongest privacy option is self-hosting: prompts never leave infrastructure you control.
Energy, carbon and water
Carbon footprint reads 49/100. The energy mix is mixed, and disclosure covers some but not all serving regions.
Water and local impact reads 41/100. Some sites use evaporative cooling in regions with moderate water stress.
Speed and efficiency also matter here: a 74/100 speed reading means fewer compute-seconds per answer for routine work.
Diagnostic notes
Open weights allow local inference; parent company scores poorly on water siting.
Readings are DoctorKnow.ai's own 0–100 assessments, compiled from vendor documentation and public reporting, and revised as new disclosures land. They are a comparison aid, not a certification.