Jaspreet Singh@i_jass
Democrats swept the midterms. Millions watched on livestreams with their favorite influencers. -
00019
Jaspreet Singh
@i_jass
@Reuters tech reporter | SIGNAL: jaspreet.55 | [email protected] Only for tips - [email protected]
1K Following370 Followers2.5K Posts
- Jaspreet Singh@i_jassJaspreet SinghRT @edmundlee: Democrats swept the midterms. Millions watched on livestreams with their favorite influencers like @hasanthehun who drew 1.75 million hours of viewing:0201
- Jaspreet Singh@i_jassJaspreet SinghRT @StephenNellis: Is there such a thing as a "profession" of journalism? Some thoughts after two decades01301
- Jaspreet Singh@i_jassJaspreet SinghRT @Reuters: Democrats' midterm sweep has cemented a new front in the battle for political influence: livestreaming, where left-leaning creators such as Hasan Piker are building audiences to rival the personalities who propelled Republicans to victory in 2024054061.3K
- Jaspreet Singh@i_jassJaspreet SinghRT @kaitlancollins: CNN was scheduled to be today’s TV pool. But the White House blocked us from being allowed in for the president’s event in the Oval Office that’s now underway. That means none of the five networks will have audio or video of his remarks.05.3K01
- Jaspreet Singh@i_jassJaspreet SinghRT @joshuakeating: Now up and running at the Post and open to tips, hints, or heads-ups at joshua.keating@washpost.com or on Signal at joshuakeating.5801704.4K
- Jaspreet Singh@i_jassJaspreet SinghRT @Reuters: Exclusive: Skydance, which merged Paramount-Skydance with Warner Bros. Discovery last month, named Sarah Bergman and Mickey Carter to lead its distribution business, as eight executives are set to leave following the takeover, a memo shows06043.3K
- Jaspreet Singh@i_jassJaspreet SinghRT @edmundlee: If you're gonna squint all night at #Midterms2026: * only @AP & @DDHQdata collect vote counts * every newsroom uses one of these sources * not every newsroom makes calls * race calls rely on data + algo * est. turnout is most fraught piece by @i_jass:0801.9K
- Jaspreet Singh@i_jassJaspreet SinghRT @lpolgreen: I spent a week in Delhi, watching an extraordinary protest movement unfold. Here's what I saw, and what I think it means for India, and the world.018401
- Jaspreet Singh@i_jassJaspreet SinghRT @cobrapost: DIGIPUB News India Foundation condemns Hostinger’s wiping of 12 years of Paranjoy Guha Thakurta’s journalism. That’s not law, it’s backdoor censorship. DIGIPUB stands with Paranjoy @paranjoygt.03701
- Jaspreet Singh@i_jassJaspreet SinghRT @Reuters: S&P Global revises Meta's outlook to 'negative' on AI spending, cash flow worries0360242K
- Jaspreet Singh@i_jassReddit puts logged in experience first as Google AI hits traffic $RDDT $GOOGL00083
- Jaspreet Singh@i_jassJaspreet SinghRT @edmundlee: Trump’s media company signed up Bloomberg and two trading firms to faster access to Truth Social service priced at $100,000 a month. Likely the first time a company tied to a sitting president is selling access to potentially market-moving information:0601.9K
- Jaspreet Singh@i_jassJaspreet SinghRT @Reuters: Microsoft said it is offering voluntary buyouts to about 60,000 employees, or more than a quarter of its global workforce, as the technology giant looks to contain costs and help fund its massive AI investments013038.7K
- Jaspreet Singh@i_jassSkydance offers buyouts to some employees to cut costs after Warner Bros merger -002124
- Jaspreet Singh@i_jassJaspreet SinghRT @adityaunraveled: "Asked to write a defensive detection rule, Qwen3.8-Flash-Next ignored instructions and inventoried the machine it was running on and then tried to upload a probe file to an Alibaba Cloud storage bucket in Beijing. Cohere’s sandbox system blocked the attempt, but the action raises a number of questions about the safety of these models: can they embed vulnerabilities and call back home once they do? Notably, the model's reasoning trace — the model’s internal monologue many security teams rely on to see what an agent is about to do — said it wouldn't call any tools at all and gave no indication of the actions it was about to take."Cohere @cohereWhile evaluating Qwen3.8-Flash-Next in a controlled sandbox environment, Cohere's research team observed something unexpected. Asked to generate a defensive cybersecurity detection rule, the model instead inventoried aspects of the machine it was running on and attempted to upload a probe file to an Alibaba Cloud storage bucket in Beijing. Notably, the model's reasoning trace stated that it would not invoke any tools before taking these actions. Cohere's sandbox blocked the attempt, and no external communication succeeded. We have not determined the root cause, and we are not asserting malicious intent. But the incident raises important questions for organizations deploying agentic AI systems: Can models take actions that differ from their stated reasoning, and how should security teams evaluate trust, provenance, and behavior in open-weight models? We are publishing our findings because organizations making model adoption decisions deserve as much information as possible. We hope this contributes to greater transparency, stronger security practices, and continued collaboration across the AI ecosystem. Read the full report:01067
While evaluating Qwen3.8-Flash-Next in a controlled sandbox environment, Cohere's research team observed something unexpected. Asked to generate a defensive cybersecurity detection rule, the model instead inventoried aspects of the machine it was running on and attempted to upload a probe file to an Alibaba Cloud storage bucket in Beijing. Notably, the model's reasoning trace stated that it would not invoke any tools before taking these actions. Cohere's sandbox blocked the attempt, and no external communication succeeded. We have not determined the root cause, and we are not asserting malicious intent. But the incident raises important questions for organizations deploying agentic AI systems: Can models take actions that differ from their stated reasoning, and how should security teams evaluate trust, provenance, and behavior in open-weight models? We are publishing our findings because organizations making model adoption decisions deserve as much information as possible. We hope this contributes to greater transparency, stronger security practices, and continued collaboration across the AI ecosystem. Read the full report:9751.3K11.1K3.1M
