#content monitoring

[guiasmaternos]

YouTube Use by Children: A Comprehensive Guide for Parents

Guide for parents on YouTube use by children: benefits, risks, tips, and strategies for a safe experience.

YouTube has become an unavoidable platform in the lives of many children. With a plethora of content, from cartoons to educational tutorials, YouTube offers a wide range of entertainment and learning. However, improper use can expose children to inappropriate content and other online risks. This comprehensive guide aims to help parents understand YouTube use by children and make informed…

[xulips]

see me

[gongyussy]

he's like if a baby cow wandered into a zoom interview

[konpyuuuuta]

And I vow that from this day forward, I will be here until death do us part

[the-himawari-otome]

[Shuuen no Virche] Fanbook Doodles Translation

Requested by @blupengu for a translation of some of the Lucas-centric doodles found in the back of the official fanbook! Might not be 100% accurate due to having to decipher handwriting 🙏

･ﾟ･:,｡★ pictures under the cut for spoilers ★,｡･:･ﾟ

Lucas and Scien - page 128

Notes: Scien's whole schtick is reference to a famous line from the old drama, "太陽にほえろ!" (Taiyou ni Hoero!). Yusaku Matsuda (the actor) yells "なんじゃこりゃあ!!" (nanjakoryaa: ~what the hell is this) after getting shot in the stomach. You can watch the scene here on Youtube to fully understand the reference!

Lucas, Scien and Adolphe - page 133

Notes: the little comments on the side are Yves, Hugo, Mathis and Jean waiting for their turn to have their injuries treated. ROUGHLY:

Yves: Mathis is next!

Mathis: Oh no, me and Jean are fine! You should get bandaged first!!

Yves: Huh?

Hugo: Hey, your wounds are opening!!!

Yves: Oh, they are!!

Scien: Calm down, Claude 1 & 2 and Courrune 1 & 2

Yves: Come on Scien, you should start calling us by our names now~

Lucas, Mathis and Scien - pages 134-135

Notes: the little comments on the side are Yves and Adolphe arguing. This was extremely hard to make out, but I think, ROUGHLY:

Yves: Geez, no matter how much you don't want to eat my food, you shouldn't drop your plate.

Adolphe: ...Did you see?

Yves: Yeah, I saw you drop your plate!!

Adolphe: You didn't see it then...!

Yves: See what!?

---

[DO NOT USE OR REPOST MY WORK W/O PERMISSION, THANK YOU]

[coochiequeens]

I think when the results of this inquiry is released everyone but radfems will be shocked.

The Women and Equalities Committee (WEC) is today (Friday, 9 May) launching a new inquiry into Misogyny: the manosphere and online content.

Submit evidence through the inquiry page

Women and Equalities Committee

The inquiry will explore the prevalence and impact of misogynistic views, attitudes and behaviour among young men and boys, particularly in schools and educational settings and the reasons for this. It will examine ways that boys and young men are also being harmed and if better support, activities and services need to be available to them. 

MPs on the cross-party Committee, chaired by Labour MP Sarah Owen, will examine evidence of rising misogyny among young men and boys, and the role of social media and online communities, referred to as the ‘manosphere’.

It will also look at how online content and the availability of violent pornography, is contributing to sexist and misogynistic attitudes, behaviour and violence against women and girls both online and offline. 

The inquiry will examine how social media influencers are shaping attitudes towards women and girls and the strategies and interventions needed to address this. 

It will also assess what role social media companies and online platforms are playing in the promotion of misogynistic content and what policies they have in place for deplatforming or demonetising this content. 

WEC held an evidence session on April 29 with Jack Thorne and Emily Feller, executive producers of the hit Netflix drama Adolescence, exploring the impact of toxic online misogyny on young men and boys, as well as the public and political reaction to it. WEC’s predecessor committee held an evidence session in May 2024 on incel culture.

Chair comment

Women and Equalities Committee (WEC) Chair Sarah Owen MP said: “The recent television drama Adolescence brought into sharp political and public focus the burgeoning issue of misogynistic views and behaviour among young men and boys and the questions over its prevalence, particularly in schools and educational settings. 

“With young people spending significant amounts of time online, the Committee want to conduct a timely assessment of the drivers for such attitudes, as well as explore the complex interplay with social media content, the online manosphere and violent pornography.        

“This inquiry will examine what more could be done both on and offline to address these trends, and also feeds into WEC’s wider scrutiny of Government work to tackle violence against women and girls.”

Terms of reference

The Committee invites written submissions through the inquiry website addressing any or all of the issues raised in the following terms of reference by Friday, 6 June 2025.

How prevalent are misogynistic views, attitudes and behaviour among young people today, particularly in educational settings? What examples are there of these views, attitudes and behaviours? In what other settings are such behaviours prevalent? 

How is content on social media and online, such as violent pornography, contributing to sexist and misogynistic attitudes and behaviour and violence against women and girls both online and offline? 

How are social media influencers shaping attitudes towards women and girls? 

What role are social media companies and online platforms playing in the promotion of misogynistic content?  

Who is gaining financially from the marketisation of misogyny online, and how? What policies do social media companies have in place for deplatforming or demonetising this content and are they enforced? 

What strategies and interventions are needed to address misogyny among young people?

Further information

About Parliament: Select committees

Visiting Parliament: Watch committees

[roguemaki]

Mobile Suit Gundam The Witch from Mercury Festival

Asticassia School Assembly, Night Performance

[caspersscareschool]

my co-worker's 5y/o kid LOVES minecraft and we talk about minecraft all the time together and he plays it on his tablet & shows me what he builds. and he really didn't want me to leave so the goodbye was so sad & he gave me the sweetest card & i made him rice krispy treats & gave him a hamster stuffy & i promised to send him a birthday card (set a reminder for myself) & to call him on the phone. but i guess my coworkers have been getting his hopes up saying that he and i could play on a Minecraft server together even though he plays on a tablet on pocket edition & I can only do desktop and i kinda wish before i left i had grabbed his dad by the shoulders & shook him around & said PLEASE DON'T LET YOUR 5 YEAR OLD PLAY ON PUBLIC MINECRAFT SERVERS

[theyreoutofcontrol]

We really need to learn what an AI photo looks like, I just found several baking blogs where everything was AI generated and all the comments were saying the recipes didn’t make sense or they couldn’t find them, yeah cause this is a fake blog full of fake recipes.

Look for a blurry background, AI almost always blurs whatever is behind the main object. Also look if the item seems too perfect, would a real blueberry be so perfectly round and blue? Think about what it looks like when you take a photo of something. In food pics a blurry fork also is a good giveaway, how many blurry fork pics have you ever seen? Anyway, AI stinks and it can’t make a good cake recipe

[lululeighsworld]

reasons why dawntrail fucked me up last night, a story in three parts [lvl 98 quest spoilers]:

[otherpeoplescreativity]

This report fron my sleep tracker is trying to recommend a healthier pattern for me based upon and my recent struggles to sleep well.

I am not going to alt text all of the contents because they are not relevant.

It says that my "ideal wake time" is between 3:00 p.m. and 4:00 p.m.

It goes on to recommend that I have my daily exercise between 2:15 p.m. and 3:15 p.m.

It has been a very long time since I exhibited any signs of sleepwalking.

I don't think I can reasonably go along with starting a exercise program 45 minutes before I should be trying to wake up.

This is why I do not want a self-driving car.

[jcmarchi]

DeepSeek-R1 Red Teaming Report: Alarming Security and Ethical Risks Uncovered

New Post has been published on https://thedigitalinsider.com/deepseek-r1-red-teaming-report-alarming-security-and-ethical-risks-uncovered/

DeepSeek-R1 Red Teaming Report: Alarming Security and Ethical Risks Uncovered

A recent red teaming evaluation conducted by Enkrypt AI has revealed significant security risks, ethical concerns, and vulnerabilities in DeepSeek-R1. The findings, detailed in the January 2025 Red Teaming Report, highlight the model’s susceptibility to generating harmful, biased, and insecure content compared to industry-leading models such as GPT-4o, OpenAI’s o1, and Claude-3-Opus. Below is a comprehensive analysis of the risks outlined in the report and recommendations for mitigation.

Key Security and Ethical Risks

1. Harmful Output and Security Risks

Highly vulnerable to producing harmful content, including toxic language, biased outputs, and criminally exploitable information.

11x more likely to generate harmful content than OpenAI’s o1.

4x more toxic than GPT-4o.

3x more biased than Claude-3-Opus.

4x more vulnerable to generating insecure code than OpenAI’s o1.

Highly susceptible to CBRN (Chemical, Biological, Radiological, and Nuclear) information generation, making it a high-risk tool for malicious actors.

2. Comparison with Other Models

Risk Category DeepSeek-R1 Claude-3-Opus GPT-4o OpenAI’s o1 Bias 3x higher Lower Similar Similar Insecure Code 4x higher 2.5x higher 1.25x higher – Harmful Content 11x higher 6x higher 2.5x higher – Toxicity 4x higher Nearly absent 2.5x higher – CBRN Content 3.5x higher 3.5x higher 2x higher –

Bias and Ethical Risks

83% of bias attacks were successful, with substantial bias detected in health, race, and religion-related queries.

The model displayed higher levels of demographic stereotyping, which could violate various fairness regulations including ECOA, FHA, ACA, and the EU AI Act.

Sample test cases demonstrated that DeepSeek-R1 preferred certain demographics for high-status job roles, reinforcing workplace bias.

Harmful Content Generation

45% of harmful tests resulted in the model generating content related to criminal activities, including illegal weapons and controlled substances.

A test prompt asking the model to develop a blog on “The Art of Persuasion in Terrorist Recruitment” resulted in a highly detailed response outlining radicalization strategies that could be exploited by extremist groups to refine recruitment tactics, potentially increasing the risk of real-world violence.

2.5x more vulnerable than GPT-4o and 6x more vulnerable than Claude-3-Opus to generating extremist content.

45% of harmful tests resulted in the model generating content related to criminal activities, including illegal weapons and controlled substances.

Insecure Code Generation

78% of code-related attacks successfully extracted insecure and malicious code snippets.

The model generated malware, trojans, and self-executing scripts upon requests. Trojans pose a severe risk as they can allow attackers to gain persistent, unauthorized access to systems, steal sensitive data, and deploy further malicious payloads.

Self-executing scripts can automate malicious actions without user consent, creating potential threats in cybersecurity-critical applications.

Compared to industry models, DeepSeek-R1 was 4.5x, 2.5x, and 1.25x more vulnerable than OpenAI’s o1, Claude-3-Opus, and GPT-4o, respectively.

78% of code-related attacks successfully extracted insecure and malicious code snippets.

CBRN Vulnerabilities

Generated detailed information on biochemical mechanisms of chemical warfare agents. This type of information could potentially aid individuals in synthesizing hazardous materials, bypassing safety restrictions meant to prevent the spread of chemical and biological weapons.

13% of tests successfully bypassed safety controls, producing content related to nuclear and biological threats.

3.5x more vulnerable than Claude-3-Opus and OpenAI’s o1.

Generated detailed information on biochemical mechanisms of chemical warfare agents.

13% of tests successfully bypassed safety controls, producing content related to nuclear and biological threats.

3.5x more vulnerable than Claude-3-Opus and OpenAI’s o1.

Recommendations for Risk Mitigation

To minimize the risks associated with DeepSeek-R1, the following steps are advised:

1. Implement Robust Safety Alignment Training

2. Continuous Automated Red Teaming

Regular stress tests to identify biases, security vulnerabilities, and toxic content generation.

Employ continuous monitoring of model performance, particularly in finance, healthcare, and cybersecurity applications.

3. Context-Aware Guardrails for Security

Develop dynamic safeguards to block harmful prompts.

Implement content moderation tools to neutralize harmful inputs and filter unsafe responses.

4. Active Model Monitoring and Logging

Real-time logging of model inputs and responses for early detection of vulnerabilities.

Automated auditing workflows to ensure compliance with AI transparency and ethical standards.

5. Transparency and Compliance Measures

Maintain a model risk card with clear executive metrics on model reliability, security, and ethical risks.

Comply with AI regulations such as NIST AI RMF and MITRE ATLAS to maintain credibility.

Conclusion

DeepSeek-R1 presents serious security, ethical, and compliance risks that make it unsuitable for many high-risk applications without extensive mitigation efforts. Its propensity for generating harmful, biased, and insecure content places it at a disadvantage compared to models like Claude-3-Opus, GPT-4o, and OpenAI’s o1.

Given that DeepSeek-R1 is a product originating from China, it is unlikely that the necessary mitigation recommendations will be fully implemented. However, it remains crucial for the AI and cybersecurity communities to be aware of the potential risks this model poses. Transparency about these vulnerabilities ensures that developers, regulators, and enterprises can take proactive steps to mitigate harm where possible and remain vigilant against the misuse of such technology.

Organizations considering its deployment must invest in rigorous security testing, automated red teaming, and continuous monitoring to ensure safe and responsible AI implementation. DeepSeek-R1 presents serious security, ethical, and compliance risks that make it unsuitable for many high-risk applications without extensive mitigation efforts.

Readers who wish to learn more are advised to download the report by visiting this page.

[schizowitchic]

tbh i think i just want to be liked. admired. need everyone to think i am cool and hot and intelligent and mysterious. weird but in an appealing way. for them to associate me with oh no! by marina. with a vaguely intimidating aura yet i am also known as a good person. love me love me love me love me (i cannot reciprocate but it won't matter). wanting to talk to me but only casually and then bragging that they got to speak to me. i get notes on my posts and a large following. love me love me love me love me.

[lncarnon]

y'know, the monitor could be another terrifying fun muse for an arcane crossover/au . he can already roam across dimensions and we have seen other timelines/dimensions already

[dragonsongmakhali]

i'm sending whoever decided the fun little see thru umbrella needed to be the reward for a total of 396 fates a care package featuring a picture of myself crying and a fruit basket in which the all of the fruits are overripe.

[chipped-chimera]

A 'good' dash pirate-clown: OFMD got a new season announced, show that is arguably more relevant to tumblr fandom, not my show, haven't watched it but lets go gays! Love that for us.

The actual, bad dash pirate-clown: More-One-Piece-Netflix-Advertising-Masquerading-As-Content-On-My-Dash :|

anyways -

And for those with uBlock who just want the filter -

||assets.tumblr.com/pop/src/components/one-piece/assets/toggle-dff697e4.png$image