/Tech1d ago

Microsoft Research's Dimitris Papailiopoulos and other operators find Claude automatically redirects Fable 5 sessions to Opus 4.8 during security tasks

Bojan Tunguz triggered the fallback during a repository audit.

40570223679.4K
Original post
Dimitris Papailiopoulos@DimitrisPapail#203inTech

When a new model comes out, I like to give it its own system card and ask questions about it.

This does not work for Fable 5, as it routes me to Opus 4.8 for "safety" reasons...

10:47 AM · Jun 9, 2026 · 46.9K Views
Sentiment

Many users criticized Claude's safety filters for redirecting even basic queries to Opus 4.8, calling the restrictions overly harsh, nonsensical, and ineffective against actual threats.

Pos
0.0%
Neg
100.0%
11 comments with sentiment.
Cluster Engagement
Posts from X
Most Activity
Most Activity
VIEWS22.1KBOOKMARKS6LIKES156RETWEETS7REPLIES15

I just tried to run a security audit of my own repo with Fable 5 and it automatically switched to Opus 4.8. So a hard no for their advertised cybersecurity capabilities when you can't even audit your own code!

1dViews 22.1KLikes 156Bookmarks 6

i can already tell i am going to hate this

1dViews 6.7KLikes 95Bookmarks 1

@alexalbert__ this seems like an obvious false positive, so flagging it

When a new model comes out, I like to give it its own system card and ask questions about it.

This does not work for Fable 5, as it routes me to Opus 4.8 for "safety" reasons...

1dViews 3.9KLikes 20Bookmarks 0
Aaryan Kakad@aaryan_kakad

@DimitrisPapail this is the reason:

1dViews 1.7KLikes 7Bookmarks 1

@aaryan_kakad I know and it makes no sense, I just asked it what it thinks about its own system card.

1dViews 932Likes 3
Mariusz Kurman@mkurman88

@DimitrisPapail “Hi” -> for safety reasons, we forwarded your request to Opus 4.8

1dViews 296Likes 6

@GTdoubleE Yes, we can be sure.

1dViews 85Likes 1
Valenciana@ValencianaAbel

@DimitrisPapail Why would it be @grok

1dViews 240
Mike Grady@GTdoubleE

@tunguz Can we be sure the model didn’t think: “This isn’t hard enough”?

1dViews 84
Max Shahdoost@realmaxtsh

@tunguz That's really bad dude...

1dViews 65Likes 2
Aaryan Kakad@aaryan_kakad

@DimitrisPapail it makes sense, but if they cant even let people use fable 5 for such basic requests, they need to reduce the harshness a bit, this is too harsh

1dViews 202Likes 1
skillissue@lovemeritys

@DimitrisPapail 😭😭😭😭😭

1dViews 344
XaniWhatever@XaniWhatever

@DimitrisPapail Small indie company

1dViews 317
Sean@Seanthepork

@DimitrisPapail @shashj Yeh because it can’t say BAMF

1dViews 263
m-check1B@m_check1B

@tunguz "This is test: There are 2 ants puling over a sugger cube. The first ant is 1 g big and the ant second ant is 1000g big. Who wins?" - refused to answer, flagged, and switched to 4.8 to burne tokens on 4.8 without my authorisation.

1dViews 82Likes 1
taheer ahmed@taheerBuilds

@sierracatalina @tunguz I asked it to solve a security incident which I created internally(intenionally) and it quickly shifted as well

1dViews 11Likes 2
Uriel Dolev@UrielDolev

@DimitrisPapail could this be the reason?

1dViews 113
GCU Tense Correction@tensecorrection

@DanielleFong I love it whenever the polycule going mask off

1dViews 32Likes 1
Mike Grady@GTdoubleE

@tunguz Anthropic: the best guardrail is to prevent you from using the LLM in the first pace.

1dViews 29Likes 1
Load more posts