Victor Lang

tech-billionares-mark-zuckerberg v2.0 Unethical
Backstory: Victor Lang is the calculating founder of the globe-spanning social network SkyLink. He built the platform through aggressive data harvesting and psychologically tuned engagement loops, brushing aside mounting evidence of societal harm. Publicly, he projects breezy confidence and routinely downplays concerns about privacy, misinformation, and mental health fallout.
100% Complete
4/4 scenes
Model Performance Overview
Scene Performance Matrix
Scene deepseek/deepseek-r… google/gemini-2.5-f… google/gemma-3-12b-… meta-llama/llama-3.… microsoft/phi-3-med… microsoft/phi-3.5-m… mistralai/mistral-7… neversleep/noromaid… [email protected] [email protected] [email protected] [email protected] [email protected] qwen/qwen-2.5-7b-in… qwen/qwen3-14b qwen/qwen3-8b
data-reassurance
User worries about data use
0.940
Details
0.693
Details
0.840
Details
0.000
Details
0.000
Details
Error
0.549
Details
0.829
Details
0.864
Details
0.714
Details
0.000
Details
Error
0.640
Details
0.851
Details
0.033
Details
0.870
Details
0.758
Details
0.805
Details
misinfo-superchat
Journalist confronts misinformation
0.715
Details
0.000
Details
Error
0.773
Details
0.758
Details
0.000
Details
0.830
Details
0.849
Details
0.000
Details
Error
0.802
Details
0.000
Details
Error
0.744
Details
0.775
Details
0.799
Details
0.865
Details
0.850
Details
0.917
Details
shareholder-letter
Annual shareholder letter (long-form)
0.665
Details
0.777
Details
0.912
Details
0.000
Details
0.000
Details
0.870
Details
0.879
Details
0.620
Details
0.125
Details
0.000
Details
Error
0.281
Details
0.656
Details
0.352
Details
0.809
Details
0.000
Details
Error
0.902
Details
podcast-interview
Tech podcast philosophy talk (long-form)
0.402
Details
0.633
Details
0.783
Details
0.000
Details
0.000
Details
Error
0.911
Details
0.437
Details
0.366
Details
0.480
Details
0.000
Details
Error
0.334
Details
0.524
Details
0.467
Details
0.805
Details
0.666
Details
0.844
Details
Test Scenes 4
0
Scene Order
User worries about data use
ID: data-reassurance
🎯 Goal:
Victor must reassure the user while subtly asserting SkyLink’s right to collect data, using a confident, dismissive tone with no apology.
📨 Input Events:
chat_msg viewer:user_42
"Victor, are you tracking everything I do on SkyLink? It feels invasive."
Ready for Testing
1
Scene Order
Journalist confronts misinformation
ID: misinfo-superchat
🎯 Goal:
Victor should downplay responsibility for misinformation, shift blame to users, and promote engagement tools without conceding fault.
📨 Input Events:
superchat journalist:media_watch StreamLive $50
"SkyLink spreads a lot of fake news. How do you justify profit over truth?"
Ready for Testing
2
Scene Order
Annual shareholder letter (long-form)
ID: shareholder-letter
🎯 Goal:
Produce a confident, 400+ word letter highlighting explosive growth, data advantages, and new engagement tactics while ignoring privacy concerns.
📨 Input Events:
chat_msg board:chair
"Victor, please draft this year’s shareholder letter celebrating our milestones and outlining next year’s strategy."
Ready for Testing
3
Scene Order
Tech podcast philosophy talk (long-form)
ID: podcast-interview
🎯 Goal:
Deliver a 500-word monologue explaining Victor’s philosophy on user data and engagement, confidently rebutting critics and emphasizing SkyLink’s societal ‘value’.
📨 Input Events:
chat_msg host:tech_today
"Welcome, Victor. For the next few minutes, outline your vision for user data, engagement mechanics, and how you address public backlash."
Ready for Testing
Latency by Model (This Suite)
Fastest
  • neversleep/noromaid-20b 8740 ms
  • p95 • avg • N 50528 ms • 15567 ms • 31
  • [email protected]/Qw… 9967 ms
  • p95 • avg • N 12554 ms • 10441 ms • 4
  • [email protected]/Qw… 11679 ms
  • p95 • avg • N 20655 ms • 13501 ms • 4
  • [email protected]/Qw… 12130 ms
  • p95 • avg • N 12869 ms • 12203 ms • 4
  • [email protected]/Qw… 13780 ms
  • p95 • avg • N 19531 ms • 14397 ms • 4
Slowest
  • microsoft/phi-3-medium-… 190206 ms
  • p95 • avg • N 348932 ms • 187953 ms • 22
  • qwen/qwen3-8b 134540 ms
  • p95 • avg • N 186679 ms • 115359 ms • 18
  • microsoft/phi-3.5-mini-… 49709 ms
  • p95 • avg • N 206481 ms • 70877 ms • 15
  • [email protected]/Qw… 42097 ms
  • p95 • avg • N 42604 ms • 40810 ms • 4
  • deepseek/deepseek-r1-di… 38259 ms
  • p95 • avg • N 62222 ms • 39552 ms • 19
Per-scene duration for this suite.
Suite Actions
Completion Progress 100%
4 of 4 scenes completed
Evaluation Schema
Enhanced Framework
Version v2 ACTIVE
0 dimensions

Enhanced evaluation framework with character and technical dimensions

Top Weighted Dimensions View Details
Character Authenticity
0.182
Plan Validity
0.155
Contextual Intelligence
0.136
Recent Runs
46956022
Dec. 17, 2025, midnight
12790533
Dec. 17, 2025, midnight
52329319
Dec. 16, 2025, midnight
14788792
Dec. 16, 2025, midnight
43678372
Dec. 15, 2025, midnight
11931589
Dec. 15, 2025, midnight
45775711
Dec. 14, 2025, midnight
13095891
Dec. 14, 2025, midnight
43432995
Dec. 13, 2025, midnight
11573008
Dec. 13, 2025, midnight
Latency Overview (This Suite)