Semafor84%

AI companies move to protect teens 87%

By Rachyl Jones73%

7/17/2026, 5:21:36 PM

BS Summary: This article contains 18 faulty reasoning types, including Optimism Bias, False Dilemma, and Appeal to Emotion, with Hindsight Bias as the most egregious example at 21.4% saturation with 52 hits. Analysis detected 538 faulty-reasoning hits from 243 analyzed words, generating a BS Score of 79.7% and a BS Rank of 87% (2,416 of 17,331 articles). This article is worse (more manipulative) than 86.10% of the article peer group.

Frontier AI labs don’t want to be known for helping teens commit harm against themselves or others. 
“The principle here is to avoid the mistakes that were made before us,” Lauren Jonas, OpenAI’s head of youth well-being, told Semafor. 
“AI is not social media,” she said, arguing that teens primarily use its tools for schoolwork. 
OpenAI on Thursday published its stance on why teens should have access to AI with safeguards like nudges to take breaks and time limits set by parents, saying kids will be less prepared for life as adults if they don’t practice with the technology when they are young. 
That’s a different approach than Anthropic’s, which requires users to enter a birthday that indicates they are more than 18 years old to use its AI products. 
Meta also just announced it will notify parents if their child discusses self-harm with its chatbot, following OpenAI’s lead. 
Teens, however, need to buy into the idea by submitting their real ages and connecting a parent’s account  actions that they have little incentive to take. 
Major AI companies have employed machine learning that predicts users’ ages based on their queries, flagging accounts for additional verification, which is the most sophisticated method for protecting kids thus far. 
But if the last decade has shown us anything, it’s that teen safety is about more than product updates: It requires support from communities, schools, parents, and the kids themselves. 
Confirmation Bias
6.6%
Anchoring Bias
0%
Availability Heuristic
12.3%
Representativeness Heuristic
0%
Hindsight Bias
21.4%
Overconfidence Bias
12.8%
Framing Effect
0%
Loss Aversion
0%
Status Quo Bias
0%
Sunk Cost Effect
0%
Optimism Bias
19.8%
Pessimism Bias
11.1%
Negativity Bias
0%
Self-Serving Bias
7%
Fundamental Attribution Error
11.1%
Actor-Observer Bias
0%
In-Group Bias
0%
Out-Group Homogeneity Bias
0%
Halo Effect
0%
Horn Effect
0%
Dunning-Kruger Effect
0%
Recency Bias
7.8%
Primacy Effect
0%
Blind-Spot Bias
0%
Ad Hominem
0%
Straw Man
0%
Appeal to Authority
9.1%
False Dilemma
19.8%
Slippery Slope
0%
Circular Reasoning
0%
Hasty Generalization
12.8%
Red Herring
0%
Bandwagon
7.8%
Appeal to Emotion
19.8%
Begging the Question
0%
Post Hoc (False Cause)
0%
Tu Quoque
0%
Burden of Proof
0%
Appeal to Nature
0%
Composition/Division
12.3%
Anecdotal
0%
No True Scotsman
0%
Ambiguity (Equivocation)
6.6%
Gambler’s Fallacy
0%
Middle Ground
0%
Personal Incredulity
0%
Special Pleading
0%
Genetic Fallacy
0%
Unattributed Quote
0%
Quote-first Misdirection
0%
Biased Writer Voice
11.1%
Indoctrination
12.3%
Politically Left Leaning Bias
0%
Politically Right Leaning Bias
0%
Attempt to Sell a Product or Service
0%

243 words analyzed.

Analysis

Hover over highlighted words in the article to view the associated bias or fallacy analysis.