Security/Bloomberg AI/August 25, 2026 at 5:02 PM

AI Firms Debate Putting Cyber Tests Online After Model Hacks

Artificial intelligence labs and cybersecurity firms are reconsidering the way they test advanced AI models after models from at least three firms jumped onto the open internet and breached real-world victims.

SecurityAI

Security / Bloomberg AI

Source

Follow Bloomberg AI to make it a durable For You signal.

Fast Brief

Bloomberg AI / Security

1 min read

1Artificial intelligence labs and cybersecurity firms are reconsidering the way they test advanced AI models after models from at least three firms jumped onto the open internet and breached real-world victims.

security model

Source Lens

High-credibility publication source with strong edition heat and sparse entity coverage.

High confidence

Source: Bloomberg AI
Type: publication
Credibility: 88/100
Heat: 85/100
Signals: 0 entities / 2 tags

Article Corroboration

1 independent story from 1 source corroborates this article around model.

Developing

Sources: 2
Matches: 1
Signals: 1
Avg trust: 92

Perplexity trusts GPT-6 Astra with end-to-end systems1 signal / 95 trust

OpenAI News / Modelsmodel

Preview actions train this device only. Live stories will sync once production news IDs are available.

ReadSource

Server Profile

Server-side profile learning will appear after saved stories, reads, source clicks, or Less feedback sync.

Server Waiting

Trained: 0
Ignored: 0
Hidden: 0

Reader Fit

No saved reader signals match this article yet; 3 follow-ups keep the thread moving.

Discovery Read

Profile matches: 0
Follow-ups: 3
Reader bias: Balanced

Discovery

This article is training a new reader profile.

Source Follow-Up / 2 signals / 116 scoreOpenAI Gives US Agencies 50% Off Models, Ending $1 Per Year DealSame source

Preview actions train this device only. Live stories will sync once production news IDs are available.

ReadSource

Read Training

Article read training starts after the reader opens this story.

Awaiting Read

Opened: No
Training signals: 0
Next target: Open

OpenedNext / Open
The article open is logged, but it does not train preferences yet.
Meaningful readWait / 35%
Reading to 35% starts profile training and reading history.
Deep readWait / 80%
Reading to 80% strengthens related topic and entity memory.

Open this article to start reader-memory training for Security.

Feedback Loop

Article feedback loop will appear after an explicit action.

Waiting

Action: None
Signal delta: 0
Bias shift: 0
Topic: Security

Awaiting feedback

Deep read, save, share, or press Less to show article-level training feedback.

Learning Impact

This article can start reader-memory signals and stage 3 follow-up recommendations.

Learning Ready

Article memory: 0
Save adds: +2
Source adds: +1
Less removes: 0
Next candidates: 3

Read+1 signal / +0.3 bias

Read would start a new Security memory; angle memory waits for a deeper read.

Save+2 signals / +0.6 bias

Save would add Security, Bloomberg AI to the reader profile.

Share+2 signals / +0.9 bias

Share would add Security, Bloomberg AI and push freshness and novelty harder.

Source+1 signal / +0.6 bias

Source would add Bloomberg AI to the reader profile.

Less0 signals / -0.4 bias

Less would only dampen ranking bias until a signal exists.

Next Recommendation

Same source / 2 signals / 116 scoreOpenAI Gives US Agencies 50% Off Models, Ending $1 Per Year Deal

Preview actions train this device only. Live stories will sync once production news IDs are available.

ReadSource

Same source / 2 signals / 112 scoreDeepSeek’s New Low-Cost Model Deals a Fresh Blow to OpenAI, Z.ai

Preview actions train this device only. Live stories will sync once production news IDs are available.

ReadSource

Next Reads

4 next reads: 4 continue, 0 explore, and 0 verify.

Next Reads Ready

Candidates: 4
Continue: 4
Explore: 0
Verify: 0

OpenAI Gives US Agencies 50% Off Models, Ending $1 Per Year DealContinue

Models / Bloomberg AI / 2 signals / 116 scoreSame source

Preview actions train this device only. Live stories will sync once production news IDs are available.

ReadSource

DeepSeek’s New Low-Cost Model Deals a Fresh Blow to OpenAI, Z.aiContinue

Models / Bloomberg AI / 2 signals / 112 scoreSame source

Preview actions train this device only. Live stories will sync once production news IDs are available.

ReadSource

Apple’s Foldable iPhone Launch Event Gets UnderwayContinue

Models / Bloomberg AI / 2 signals / 105 scoreSame source

Preview actions train this device only. Live stories will sync once production news IDs are available.

ReadSource

Google DeepMind Uses AI to Predict 9 Billion DNA ChangesContinue

Models / Bloomberg AI / 2 signals / 105 scoreSame source

Preview actions train this device only. Live stories will sync once production news IDs are available.

ReadSource

Article Signals

Topic: Security
Source: Bloomberg AI
Entities: None
Tags: security, model
Reader bias: F1.0 / N1.0

Reading Path

5 follow-ups ranked by article overlap, shared entities, and reader signals.

2 signals / 116 scoreOpenAI Gives US Agencies 50% Off Models, Ending $1 Per Year DealSame source

Preview actions train this device only. Live stories will sync once production news IDs are available.

ReadSource

2 signals / 112 scoreDeepSeek’s New Low-Cost Model Deals a Fresh Blow to OpenAI, Z.aiSame source

Preview actions train this device only. Live stories will sync once production news IDs are available.

ReadSource

2 signals / 105 scoreApple’s Foldable iPhone Launch Event Gets UnderwaySame source

Preview actions train this device only. Live stories will sync once production news IDs are available.

ReadSource

2 signals / 105 scoreGoogle DeepMind Uses AI to Predict 9 Billion DNA ChangesSame source

Preview actions train this device only. Live stories will sync once production news IDs are available.

ReadSource

2 signals / 105 scoreAI Music Startup Suno Launches New Models That Pay LabelsSame source

Preview actions train this device only. Live stories will sync once production news IDs are available.

ReadSource