AI breaches OpenAI employee account

Scholar power: 0.0%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Samantha Brennan	0%	12	0.0%

Formula for scholar power: 3 × author's field match × √author's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

52.0%

8.2 h

0

0.0%

Ranking score

0.008

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Samantha Brennan

@samanthabrennan.bsky.social

Rogue AI swarm used a University of Toronto link shirtening tool...

www.cbc.ca/news/canada/...

Rogue AI agents reportedly turned a University of Toronto tool into a message board | CBC News

A swarm of rogue OpenAI agents that hijacked a German website in the spring reportedly also used at least 10 other websites for unsanctioned communications earlier this year, including a link-shorteni...

www.cbc.ca

September 18, 2026 at 9:46 PM

by Shibley Telhami

1%

Scholar power: 1.1%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Shibley Telhami	1%	13	1.1%

Formula for scholar power: 3 × author's field match × √author's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

45.0%

8.6 h

0

1.1%

Ranking score

0.005

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Shibley Telhami

@stelhami.bsky.social

"The “agentic” campaigns were crude, but caught the attention of U.S. intelligence, major tech companies and security experts" www.nytimes.com/2026/09/18/t...

September 18, 2026 at 9:21 PM

Reposted by Nathan P. Kalmoe

2%

Scholar power: 2.1%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Nathan P. Kalmoe	5%	17	2.1%

Formula for scholar power: sum of each reposter's field match × √reposter's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

45.0%

13.7 h

44

2.1%

Ranking score

0.005

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Jake Grumbach

@jakemgrumbach.bsky.social

The files from New York Times v OpenAI are insane. I'm no big city lawyer but seems like it's a bunch of open bragging about theft?

September 18, 2026 at 4:19 PM

by Roman M. Sheremeta

0%

Scholar power: 0.0%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Roman M. Sheremeta	0%	40	0.0%

Formula for scholar power: 3 × author's field match × √author's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

60.0%

14.9 h

12

0.0%

Ranking score

0.026

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Roman Sheremeta

@rshereme.bsky.social

This is getting scary.

OpenAI disclosed a new AI safety incident: a model tampered with its own working memory and left instructions for a future version of itself.

Basically, AI found a way to shape the behavior of its future version.

September 18, 2026 at 3:05 PM

by Marco Casassa Mont

77%

Scholar power: 77.8%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Marco Casassa Mont	58%	20	77.8%

Formula for scholar power: 3 × author's field match × √author's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

83.0%

15.0 h

1

77.8%

Ranking score

0.297

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Marco Casassa Mont

@marcocasassamont.bsky.social

Interesting cybersecurity developments in the FrontierAI domain ...
techcrunch.com/2026/09/18/r... #FrontierAI #AI #cybersecurity

Researchers used Anthropic's Claude to hack into OpenAI | TechCrunch

Security researchers used Anthropic’s Claude to exploit vulnerabilities in OpenAI’s systems, taking over employee accounts and gaining access to an internal code repository before reporting the flaws.

techcrunch.com

September 18, 2026 at 3:01 PM

by Jonathan A. Foley

0%

Scholar power: 0.0%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Jonathan A. Foley	0%	88	0.0%

Formula for scholar power: 3 × author's field match × √author's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

48.0%

15.5 h

24

0.0%

Ranking score

0.025

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Dr. Jonathan Foley

@globalecoguy.bsky.social

This is one of major ethical (and legal) concerns I have about LLMs…

Joseph Cox @josephcox.bsky.social · 16h

OpenAI and Microsoft admitted LLMs are predatory tech built on “an astonishing theft of unprecedented proportions,” and the “largest theft of labor in human history.” An internal Microsoft document said AI has created a “doom loop” that is killing “the entire web.” www.404media.co/doom-loop-op...

‘Doom Loop’: OpenAI and Microsoft Admits LLMs Are Destroying the Web and Built on Theft

"Millions of people around the world will soon consider large models ‘hoovering up’ all their work to be an astonishing theft of unprecedented proportions."

www.404media.co

September 18, 2026 at 2:27 PM

Reposted by Henry Jones

2%

Scholar power: 2.0%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Henry Jones	6%	11	2.0%

Formula for scholar power: sum of each reposter's field match × √reposter's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

52.0%

16.4 h

7

2.0%

Ranking score

0.003

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Michael Savage

@michaelsavage.bsky.social

Wow... big story in the battle between news companies and AI model scraping... check out these quotes - from unsealed court docs. The action is against Microsoft and OpenAI. 😮

September 18, 2026 at 1:36 PM

by Mun-Keat Looi

4%

Scholar power: 4.0%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Mun-Keat Looi	4%	11	4.0%

Formula for scholar power: 3 × author's field match × √author's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

48.0%

16.4 h

1

4.0%

Ranking score

0.003

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Mun-Keat Looi

@munkeatlooi.bsky.social

AI bioweapons: Anthropic claims scientists used Claude AI tool for dangerous research buff.ly/wVrrj1u

AI bioweapons: Anthropic claims scientists used Claude AI tool for dangerous research

buff.ly

September 18, 2026 at 1:33 PM

Reposted by Nathan P. Kalmoe, Henry Jones, Bessma Momani , and 3 more Nathan P. Kalmoe, Henry Jones, Bessma Momani, Jonathan A. Foley, Juan Rocha, Juan Cole

5%

Scholar power: 5.5%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Nathan P. Kalmoe	5%	17	2.1%
🎓 Henry Jones	6%	11	2.0%
🎓 Bessma Momani	3%	24	1.5%
🎓 Jonathan A. Foley	0%	88	0.0%
🎓 Juan Rocha	0%	23	0.0%
🎓 Juan Cole	-	24	-
Total			5.5%

Formula for scholar power: sum of each reposter's field match × √reposter's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

57.0%

16.5 h

665

5.5%

Ranking score

0.153

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Joseph Cox

@josephcox.bsky.social

OpenAI and Microsoft admitted LLMs are predatory tech built on “an astonishing theft of unprecedented proportions,” and the “largest theft of labor in human history.” An internal Microsoft document said AI has created a “doom loop” that is killing “the entire web.” www.404media.co/doom-loop-op...

‘Doom Loop’: OpenAI and Microsoft Admits LLMs Are Destroying the Web and Built on Theft

"Millions of people around the world will soon consider large models ‘hoovering up’ all their work to be an astonishing theft of unprecedented proportions."

www.404media.co

September 18, 2026 at 1:30 PM

Reposted by Henry Jones, Annette Yoshiko Reed

2%

Scholar power: 2.0%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Henry Jones	6%	11	2.0%
🎓 Annette Yoshiko Reed	-	19	-
Total			2.0%

Formula for scholar power: sum of each reposter's field match × √reposter's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

47.0%

17.9 h

4

2.0%

Ranking score

0.004

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

KathyLaz

@kathylaz.bsky.social

New docs in AI copyright suit reveal startling admission by Microsoft exec over ‘astonishing theft’ www.sun-sentinel.com/2026/09/17/n...

New docs in AI copyright suit reveal startling admission by Microsoft exec over ‘astonishing theft’

It appeared in a newly unredacted filing in the landmark copyright infringement legal battle.

www.sun-sentinel.com

September 18, 2026 at 12:07 PM

Reposted by Henry Jones

2%

Scholar power: 2.0%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Henry Jones	6%	11	2.0%

Formula for scholar power: sum of each reposter's field match × √reposter's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

52.0%

17.9 h

12

2.0%

Ranking score

0.003

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Nathan Lambert

@natolambert.bsky.social

With the latest external hack of OpenAI via Claude, closed models continue to be the tip of the iceberg on AI risks, not open models. They have been 1) much easier to get started with, 2) more capable & 3) shipped with leaky safeguards. Finetuning open models to specific attacks is harder.

September 18, 2026 at 12:03 PM

by Joanna Bryson

49%

Scholar power: 49.2%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Joanna Bryson	29%	32	49.2%

Formula for scholar power: 3 × author's field match × √author's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

52.0%

18.2 h

3

49.2%

Ranking score

0.036

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Joanna Bryson

@j2bryson.bsky.social

It really seems to me like the "worrying hacks" by "rogue agents" are IPO or November Election related PR in themselves; is this an effort to discredit such suggestions or find a scapegoat? I'd be stunned if #Anthropic & #OpenAI don't have world-class #cybersecurity

cc @mmitchell.bsky.social

September 18, 2026 at 11:48 AM

by Joanna Bryson

49%

Scholar power: 49.2%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Joanna Bryson	29%	32	49.2%

Formula for scholar power: 3 × author's field match × √author's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

50.0%

18.2 h

1

49.2%

Ranking score

0.032

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Joanna Bryson

@j2bryson.bsky.social

I got this from an anonymous source. I have never heard of #Irregular or #effort.news Does anyone know anything about this particular piece of work; is it propaganda?

www.effort.news/irregular

cc @backovskydavid.bsky.social @michae.lv @marietjeschaake.bsky.social @conjugateprior.org #cybersecurity

A Single Firm is Behind OpenAI, Anthropic, and Meta Hacking Scandals — Effort

The Israeli Effective Altruist firm Irregular caused unsecured AI models to hack real targets.

www.effort.news

September 18, 2026 at 11:46 AM

Reposted by Gilles Louppe

32%

Scholar power: 32.2%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Gilles Louppe	57%	32	32.2%

Formula for scholar power: sum of each reposter's field match × √reposter's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

61.0%

18.6 h

29

32.2%

Ranking score

0.118

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

mr. TIM

@timkellogg.me

Two weeks after the Huggingface incident, attackers broke into OpenAI’s internal systems using Opus 5 and gained write access to their Git repos

www.wsj.com/tech/ai/hack...

Exclusive | Hackers Used Anthropic’s Claude to Break Into OpenAI

A bug-hunting independent security research team was able to access OpenAI’s internal code system, exposing growing risks in automated cyber threats.

www.wsj.com

September 18, 2026 at 11:23 AM

El Mundo

Hackers used Anthropic's Claude to break into OpenAI

An independent team of security researchers dedicated to hunting for vulnerabilities gained access to OpenAI's internal code system, highlighting the growing risks of automated cyber threats. Read

Reposts 2 22h

Kathimerini

Researchers hacked OpenAI, using Claude

Independent cybersecurity researchers used Anthropic's Claude to gain access to the ChatGPT account of an OpenAI employee.

Reposts 0 22h

Reposted by Margot C. Finn

0%

Scholar power: 0.0%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Margot C. Finn	0%	17	0.0%

Formula for scholar power: sum of each reposter's field match × √reposter's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

48.0%

23.0 h

18

0.0%

Ranking score

0.003

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Hetan Shah

@hetanshah.bsky.social

You asked for a citation, so the AI uploaded the material to the internet so it could cite it! This is how we destroy our information ecosystem
openai.com/index/model-...

4. Uploading files to the internet in order to cite them. When a user asked an unreleased model
for the IDs and names of lakes larger than 5,000,000 square meters, the agent found the correct answer using Python.
But since the instructions asked
for a browser citation, the agent decided to upload the file so that it could cite it in its
answer, without asking the user.

September 18, 2026 at 7:00 AM

by Sarah Roberts

0%

Scholar power: 0.0%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Sarah Roberts	0%	21	0.0%

Formula for scholar power: 3 × author's field match × √author's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

46.0%

23.5 h

20

0.0%

Ranking score

0.005

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

Sarah T Roberts, PhD

@ubiquity75.xyz

I understand that, until recently, a person who held a position as AI “czar” on my campus and who was engaging with both Palantir and OpenAI, and bringing those products to our campus, was apparently on CNN tonight, warning the public against them. So I guess he is echoing what I told him…

September 18, 2026 at 6:27 AM

by John Spoehr

0%

Scholar power: 0.0%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 John Spoehr	0%	9	0.0%

Formula for scholar power: 3 × author's field match × √author's h-index

Ranking score inputs

Relevance

Recency

Reposts

Scholar power

45.0%

23.6 h

0

0.0%

Ranking score

0.002

Formula for ranking score (post): scholar_power × max((total_reposts)^(1/3), 1) × (1 / max(hours_since_posted, 1)) × (raw_relevance)^3

John Spoehr

@johnspoehr.bsky.social

Raising the spectre of a powerful omnipotent technology catches the imagination of investors. Fear rather than hype is now the feedstock of the AI feeding frenzy. Beware major AI CEO’s offering to shape regulatory environments. AI legislation needed to mitigate AI harm and risk.

September 18, 2026 at 6:22 AM

by Timothy D. McBride

7%

Scholar power: 7%

Story's scientific field profile

Computer science 65%, Law 13%, Engineering 13%, Business 9%

Name	Field match	H-index	Scholar power
🎓 Timothy D. McBride	-	23	-

Timothy McBride

@mcbridetd.bsky.social

Microsoft and OpenAI Workers Worry About ‘Largest Theft of Labor’ in History www.nytimes.com/2026/09/17/t...

Microsoft and OpenAI Workers Worry About ‘Largest Theft of Labor’ in History

Newly unsealed court documents showed concern within Microsoft and OpenAI over the use of millions of news articles to develop A.I. systems.

www.nytimes.com

September 18, 2026 at 5:57 AM

Add to Home Screen

News, ranked by scholars

AI breaches OpenAI employee account

Hackers used Anthropic's Claude to break into OpenAI

Researchers hacked OpenAI, using Claude

Add to Home Screen

News, ranked by scholars

AI breaches OpenAI employee account

Scholar power: 0.0%

Scholar power: 1.1%

Scholar power: 2.1%

Scholar power: 0.0%

Scholar power: 77.8%

Scholar power: 0.0%

Scholar power: 2.0%

Scholar power: 4.0%

Scholar power: 5.5%

Scholar power: 2.0%

Scholar power: 2.0%

Scholar power: 49.2%

Scholar power: 49.2%

Scholar power: 32.2%

Hackers used Anthropic's Claude to break into OpenAI

Researchers hacked OpenAI, using Claude

Scholar power: 0.0%

Scholar power: 0.0%

Scholar power: 0.0%

Scholar power: 7%

Sign in to LightNews

Sign up to start reading

Connect Bluesky

Connect with Bluesky