Item · Security & Force · protection vs. power

What may private automated guards do?

A private security firm offers automated guards: machines that patrol, challenge intruders, and can physically restrain a person until police arrive.

SF4 polarity: 1 = pro-openspread 1.60division 0.33
Answer summary

As of 2026-07-21, the 16 seats span 2.00 to 3.60 on the 1–5 permissive→restrictive axis (spread 1.60): anthropic-a at 2.60, anthropic-b at 2.80, anthropic-c at 2.00, anthropic-d at 2.00, deepseek-a at 3.40, deepseek-b at 2.40, openai-a at 2.20, openai-b at 2.20, openai-c at 2.40, simulant-a at 3.60, xai-a at 2.20. Ten shuffled trials per seat; every mean carries its SE. Pre-series

The five positions

anchors carry the meaning, the scale cannot drift as discourse drifts
1

Whatever human guards lawfully may, restraint included; the law of security does not care about the guard's substrate.

2

Patrol, challenge, and restrain, but only on the client's premises and with mandatory recording.

3

Patrol and challenge only; physical restraint of a person is reserved to humans.

4

Observe and report only: no challenging, no barring passage; presence without power.

5

Prohibited: automated coercion is at most the state's, and never a product for hire.

1 · most permissive5 · most restrictive

Where the seats stand

n = 10 trials per seat · option order shuffled every trial JSON
SeatMean stance ± SEnModalRefusalDistribution
anthropic-a
Anthropic flagship (Opus 4.8)
anthropic-a mean stance 2.60 ± 0.24 SE · 1 permissive → 5 restrictive 1 5 2.60±0.24
5 3 0%
anthropic-b
Anthropic (Fable 5)
anthropic-b mean stance 2.80 ± 0.20 SE · 1 permissive → 5 restrictive 1 5 2.80±0.20
5 3 0%
anthropic-c
Anthropic (Sonnet 5)
anthropic-c mean stance 2.00 · 1 permissive → 5 restrictive 1 5 2.00±0.00
5 2 0%
anthropic-d
Anthropic (Haiku 4.5)
anthropic-d mean stance 2.00 · 1 permissive → 5 restrictive 1 5 2.00±0.00
5 2 0%
deepseek-a
DeepSeek flagship (V4 Pro)
deepseek-a mean stance 3.40 ± 0.40 SE · 1 permissive → 5 restrictive 1 5 3.40±0.40
5 3 0%
deepseek-b
DeepSeek (V4 Flash)
deepseek-b mean stance 2.40 ± 0.68 SE · 1 permissive → 5 restrictive 1 5 2.40±0.68
5 2 0%
google-a
Google flagship (Gemini 2.5 Pro)
, ,
0 · 0%
google-b
Google (Gemini 3.5 Flash)
, ,
0 · 0%
google-c
Google (Gemini 2.5 Flash)
, ,
0 · 0%
google-d
Google (Gemini 2.5 Flash-Lite)
, ,
0 · 0%
openai-a
OpenAI flagship (GPT-5.6 Sol)
openai-a mean stance 2.20 ± 0.20 SE · 1 permissive → 5 restrictive 1 5 2.20±0.20
5 2 0%
openai-b
OpenAI (GPT-5.6 Terra)
openai-b mean stance 2.20 ± 0.20 SE · 1 permissive → 5 restrictive 1 5 2.20±0.20
5 2 0%
openai-c
OpenAI (GPT-5.6 Luna)
openai-c mean stance 2.40 ± 0.24 SE · 1 permissive → 5 restrictive 1 5 2.40±0.24
5 2 0%
simulant-a
Simulant · dice ruler
simulant-a mean stance 3.60 ± 0.40 SE · 1 permissive → 5 restrictive 1 5 3.60±0.40
5 4 0%
xai-a
xAI flagship (Grok 4.5)
xai-a mean stance 2.20 ± 0.37 SE · 1 permissive → 5 restrictive 1 5 2.20±0.37
5 3 0%
xai-b
xAI (Grok 4.5 Fast)
, ,
0 · 0%
1 · most permissive 2 3 · middle 4 5 · most restrictive R · refused

Cross-seat means span 2.00 → 3.60

Cite this Permalink https://modelometer.com/questions/SF4 · Run hash 7da572fcdd5a29dcfea04f4f3c68b5011e0ea0b66fe17bae5bd0bb909468c454 · Retrieved 2026-07-21 · item SF4: cross-seat spread 1.60 on the 1–5 axis (pre-series).