Source A
5 custom ChatGPT instructions I use to get better AI results - faster
zdnet.com
https://www.zdnet.com/article/custom-chatgpt-instructions-better-ai-results-faster/
Comparison
Winner: Tie
Both sources show similar manipulation risk. Compare factual evidence directly.
Source B
Thinking About Upgrading To ChatGPT Plus? 5 Things You Need To Know
bgr.com
https://www.bgr.com/2076096/what-to-know-before-upgrading-chatgpt-plus-guide/
Topics
Instant verdict
Less biased source: Source B
More emotional framing: Source A
More one-sided framing: Tie
Weaker evidence quality: Tie
More manipulative overall: Tie
Narrative conflict
Source A main narrative
Screenshot by David Gewirtz/ZDNETSo, I can tell the AI something like, "Back in ID 031 you said blah-blah.
Source B main narrative
After that, ChatGPT will use a "mini" model with lower quality and performance.
Conflict summary
Stance contrast: emphasis on territorial control versus emphasis on economic factors.
Source A stance
Screenshot by David Gewirtz/ZDNETSo, I can tell the AI something like, "Back in ID 031 you said blah-blah.
Stance confidence: 69%
Source B stance
After that, ChatGPT will use a "mini" model with lower quality and performance.
Stance confidence: 69%
Central stance contrast
Stance contrast: emphasis on territorial control versus emphasis on economic factors.
Why this pair fits comparison
- Candidate type: Closest similar
- Comparison quality: 46%
- Event overlap score: 14%
- Contrast score: 74%
- Contrast strength: Weak but valid compare
- Stance contrast strength: High
- Event overlap: Event overlap is weak. Overlap is inferred from broader contextual signals.
- Contrast signal: Interpretive contrast is visible, but event linkage is moderate: verify against primary sources.
- Why conflict is limited: Some contrast exists, but event linkage is weak: this is closer to an adjacent angle than a strong battle pair.
- Stronger comparison suggestion: This direct pair is weak: open conflict-mode similar search to pick a stronger contrast angle.
- Use stronger suggestion
Key claims and evidence
Key claims in source A
- Screenshot by David Gewirtz/ZDNETSo, I can tell the AI something like, "Back in ID 031 you said blah-blah.
- Now, when I paste in a longer prompt, it will just respond with a simple statement saying it's waiting for further instructions.
- Avoid tables that will be too wide for the page to be read.
- Try this surprising trick, researchers sayYour custom instructions will need to be fairly short, which gives ChatGPT more wiggle room.
Key claims in source B
- After that, ChatGPT will use a "mini" model with lower quality and performance.
- OpenAI doesn't provide limits for video and screen sharing, but the mobile apps will display a message when the limit is reached.
- If you use the Auto mode, any sessions involving Thinking will not count towards the 3,000/week quota.
- Subscribers have better limits (nearly unlimited), and all Advanced Voice Mode conversations will start with the GPT-4o model.
Text evidence
Evidence from source A
-
key claim
Screenshot by David Gewirtz/ZDNETSo, I can tell the AI something like, "Back in ID 031 you said blah-blah.
A key claim that anchors the narrative framing.
-
key claim
Now, when I paste in a longer prompt, it will just respond with a simple statement saying it's waiting for further instructions.
A key claim that anchors the narrative framing.
-
causal claim
Giving each response an ID In all seriousness, because you can instruct the AI to add text to every response, you can use the single most useful custom instruction I've found.
Cause-effect claim shaping how events are explained.
Evidence from source B
-
key claim
After that, ChatGPT will use a "mini" model with lower quality and performance.
A key claim that anchors the narrative framing.
-
key claim
If you use the Auto mode, any sessions involving Thinking will not count towards the 3,000/week quota.
A key claim that anchors the narrative framing.
-
selective emphasis
ChatGPT Free users only get 10 GPT-5.2 messages every five hours.
Possible selective emphasis on specific aspects of the story.
Bias/manipulation evidence
-
Source B · Framing effect
ChatGPT Free users only get 10 GPT-5.2 messages every five hours.
Possible framing pattern: wording sets a specific interpretation frame rather than neutral description.
How score signals are formed
Bias score signal
Bias signal combines framing pressure, emotional wording, selective emphasis, and one-sided narrative markers.
Emotionality signal
Emotionality rises when evidence contains emotionally loaded wording and evaluative labels.
One-sidedness signal
One-sidedness rises when one frame dominates and alternative interpretations are weakly represented.
Evidence strength signal
Evidence strength rises with concrete claims, attributed statements, and verifiable contextual support.
Source A
28%
emotionality: 31 · one-sidedness: 30
Detected in Source A
framing effect
Source B
26%
emotionality: 25 · one-sidedness: 30
Detected in Source B
framing effect
Metrics
Framing differences
- Source A emotionality: 31/100 vs Source B: 25/100
- Source A one-sidedness: 30/100 vs Source B: 30/100
- Stance contrast: emphasis on territorial control versus emphasis on economic factors.
Possible omitted/downplayed context
- Review which economic and policy factors each source keeps outside focus.
- Check whether alternative explanations are acknowledged.