R Petras
banner
rpetras.bsky.social
R Petras
@rpetras.bsky.social
Co-founder of Psst… revolutionizing the act of speaking out to hold power to account.
https://psst.org/

Bicyclist. Traveler. My soul is in Chicago, my heart in Amsterdam and Seattle, my love in Hydra and my life in Baltimore.

Pinned
We are the Whisper Network. :-)
As Meghnad Bose, @columjournreview.bsky.social, writes:
"Enter Psst.org - a homophonically named nonprofit providing a secure depository for government employees and technology workers to share concerns about wrongdoing or mayhem."

www.cjr.org/politics/whi...
The @psst-org.bsky.social Safe offers tech workers a way to come forward with concerns discreetly. We offer pro-bono legal advice and overall support. If you are someone you know see something, you can save it in the Safe.
Psst, if you were laid off and want to speak about what you’ve seen behind the scenes at Meta, we can help.

We’ve helped lots of workers with free legal advice/support if they’re concerned about something in their current or former workplace. You don’t need to “go public.” Psst.org/safe
Safe — Psst.org
Psst.org
October 23, 2025 at 6:58 PM
Reposted by R Petras
When you talk to real tech workers instead of billionaire CEOs, they hold one common opinion about AI. 👇

We need to make it easier - and normal - for people who work in tech to be able to share their takes. A working climate that chills free speech at the peril of job loss doesn't serve anyone.
October 20, 2025 at 7:27 PM
Reposted by R Petras
Ummmm…this can’t be good.

Keeping public pressure on these companies has quite literally become an existential imperative.
New: We jailbroke four versions of ChatGPT to get instructions on how to create viruses and chemical weapons.

The instructions were all flawed, and the it didn't work on GPT-5. But 4 of the 5 big AI companies have internally found their top models are close to being able to tutor bioterrorists.
ChatGPT safety systems can be bypassed to get weapons instructions
NBC News found that OpenAI’s models repeatedly provided answers on making chemical and biological weapons.
www.nbcnews.com
October 16, 2025 at 2:07 PM
Brutal fact is that holding big tech/big AI to account really threatens the pocket book. We need to fix that - collective action within the industry can keep people safe and financially sound.
Thank you @katekenny.bsky.social for making sense of it all.
What’s the ROI on telling the truth in Big Tech/AI?

Professor @katekenny.bsky.social of @uniofgalway.bsky.social spent 15 years running the numbers—and the moral math doesn’t look great...

But she’s got ideas for changing it. 🔦

Full Q&A in our Substack (link below).
October 15, 2025 at 7:03 PM
Wrong. Accountability fosters trust. W/o trust, the military is a danger to national security.
"Oversight agencies...created a network of whistle-blowers & watchdogs that undermines the authority of commanders. Commanders who are too accountable become risk-averse."
www.nytimes.com/2025/10/07/o...
Opinion | That Hegseth Speech Was Actually Pretty Good
www.nytimes.com
October 7, 2025 at 2:34 PM
Until companies re-set their cultures to call out these concerns, we need external mechanisms, like the @psst-org.bsky.social Safe to help the public know what is going on behind the curtain.
This is concerning. The public needs guardrails for this new tech.
“…that Sora is being used for stalking and harassment will likely not be an edge case, because deepfaking yourself and others into videos is one of its core selling points.”

Far from an edge case, it’s the primary use case.
October 7, 2025 at 2:28 PM
My 86 yo mom goes to a protest every week. She has lately felt deflated. This article helped her.
October 2, 2025 at 2:38 PM
Love biking through Chicago
www.reddit.com/r/chicago/co...
From the chicago community on Reddit: Mayor Brandon Johnson on biking in Chicago
Explore this post and more from the chicago community
www.reddit.com
September 25, 2025 at 2:35 PM
Reposted by R Petras
WORKERS SPEAK OUT 🎤

Be careful who you RIF.

Aisha Coffey spent 15 years running campaigns that literally saved lives. After getting laid off by Doge, she’s leading a network of federal workers getting the truth out to the public.

Here's her “villain arc”:

psstpsst.substack.com/p/strategic-...
September 19, 2025 at 7:17 PM
Here’s the o those who don’t stay silent.
@psst-org.bsky.social
Dr. Susan Monarez: “I could have stayed silent, agreed to the demands, and no one would have known. What the public would have seen were scientists dismissed without cause and vaccine protections quietly eroded…I would have lost the one thing that cannot be replaced: My integrity.”
September 17, 2025 at 6:45 PM
Incredible hearing.
Could really benefit from more information from the inside of these companies.
@psst-org.bsky.social
Senate hearing today, "Examining the Harm of AI Chatbots."

Parents who lost children at the encouragement and instruction of an AI "friend" sounding a warning to us all.

Robbie Torney of @commonsensemedia.bsky.social on the failure of AI safety systems.

Hear what the public needs to know:
🧵⬇️
September 17, 2025 at 2:50 PM
Reposted by R Petras
Good news today. 👏👏👏

We tuned into this hearing live and shared out the most salient moments, check out our thread for a recap: bsky.app/profile/psst...
After hearing parents describe minors who faced mental health issues or died by suicide after intense hours spent with AI chatbots, lawmakers from both parties seemed to support the idea of requiring AI companies to add protections for young users.
Senators weigh regulating AI chatbots to protect kids
Parents told a Senate Judiciary subcommittee that chatbots had harmed their teens, even contributing to deaths by suicide.
www.washingtonpost.com
September 17, 2025 at 2:33 PM
Reposted by R Petras
A common thread across AI research: safeguards exist, but are way too easy to bypass. ⬇️
September 15, 2025 at 1:01 PM
More details on WhatsApp's lack of security protocols. @psst-org.bsky.social behind-the-scenes look.
Did you read the WhatsApp whistleblower story out this week?
6 security engineers for 2.5 billion users? ~100,000 hacks/day? Meta pretending the whistleblower is the problem? (surprise surprise)

Here are the juicy details the NYT left out, in our newsletter:

psstpsst.substack.com/p/a-whatsapp...
September 10, 2025 at 2:54 PM
Reposted by R Petras
For every $1 Meta has invested in the safety and security of its users, it has spent more than $7 to boost its own stock price.
September 10, 2025 at 1:14 PM
The most unsurprising statement ever.

Disparaging whistleblowers is in the PR handbook at Meta.

@psst-org.bsky.social
Dr Jason Sattizahn states that Mark Zuckerberg disparaged whistleblowers, calling past disclosures “used to construct a false narrative.” #whistleblower #Meta
September 9, 2025 at 8:20 PM
Happening now - whistleblowers in front of the Senate once again making it clear that Meta cannot be trusted.
Whistleblowers say Meta buried research showing kids — some as young as 10— faced grooming and harassment in VR. We’ll live‑update the Senate Judiciary Committee hearing as testimony unfolds. Turn on notifications for this thread. #Meta #LiveUpdates #ChildSafety #SenateHearing
September 9, 2025 at 8:15 PM
Reposted by R Petras
New Meta whistleblower says WhatsApp was shockingly unprotected: just 6 security engineers, 100,000 account hacks/day, exposed metadata. Businesses, journalists, activists, families rely on it daily.

We helped protect this worker as he brought his story to the NYT:

www.nytimes.com/2025/09/08/t...
Whistle-Blower Sues Meta Over Claims of WhatsApp Security Flaws
www.nytimes.com
September 8, 2025 at 7:56 PM
If only Meta committed to the security they claim, we wouldn't need a whistleblower to tell the truth. Time & time again, the only way the public knows what's going on is through a whistleblower. @psst-org.bsky.social helped bring this story to the public while protecting Baig.
"In an interview, Mr. Baig said that every day his team saw 'real world, actual harm happening,' such as 'account compromises, scraping impersonation, journalists being targeted.'" www.nytimes.com/2025/09/08/t...
Whistle-Blower Sues Meta Over Claims of WhatsApp Security Flaws
www.nytimes.com
September 8, 2025 at 9:15 PM
Reposted by R Petras
Whistleblowers have long been one of the most important checks on power we have. When they share, we listen.
Former infectious-diseases chief Jeanne Marrazzo and international-research chief Kathleen Neuzil allege in a whistleblower complaint they were removed from their posts after objecting to Trump administration efforts to undermine vaccines, flout court orders, withhold research money, NYT reports.
Whistle-Blower Complaints Detail Tension Over Vaccines at N.I.H.
www.nytimes.com
September 5, 2025 at 3:54 PM
Oh I will be buying more than one copy and gifting to others. Hers is a story of courage and perseverance…all to help us know truth. @reazlepuff.bsky.social is one of the main reasons I work everyday to help those who come forward with critical information.
On Sept 16, whistleblower @reazlepuff.bsky.social ‪‪publishes I Am Not Your Enemy. Prosecuted under the Espionage Act, she served the longest sentence ever for leaking docs. Her story underscores the profound personal toll of whistleblowing. #whistleblower #memoir
September 3, 2025 at 11:37 PM
It used to be that federal workers helped the federal govt do better for citizens. Now they are the enemy of the federal government. Let that sink in.

(If you have faced retaliation for reporting public concerns, @psst-org.bsky.social can give you pro bono legal advice.)
August 27, 2025 at 6:57 PM
Reposted by R Petras
Trump administration suspends FEMA employees who warned about disaster response
Trump administration suspends FEMA employees who warned about disaster response
The letter urges Congress to take action to stop the Trump administration from dismantling FEMA and prevent “another national catastrophe like Hurricane Katrina.”
buff.ly
August 27, 2025 at 6:50 PM
Critically important information for the public.
Breaking: A whistleblower alleged that DOGE put hundreds of millions of Americans at risk of being hacked by uploading Social Security info to a vulnerable cloud server.

According to him, no oversight mechanisms exist to monitor what DOGE is using the data for. 👇
www.nytimes.com/2025/08/26/u...
DOGE Put Critical Social Security Data at Risk, Whistle-Blower Says
www.nytimes.com
August 27, 2025 at 3:48 PM