DeepSeek: Free and Open-Source AI

Moved to Forum: Original Link

Cost Breakthrough: DeepSeek's R-1 matches the performance of OpenAI's o1 at just 10% of the cost, delivering high-quality AI that is both powerful and affordable.

Open-Source Advantage: Freely accessible and optimized for modest hardware, R-1 accelerates developer adoption and fuels innovation within the open-source community.

Industry Disruption: Big Tech giants like OpenAI, Google, and Meta, along with Nvidia's GPU dominance, face mounting pressure as low-cost, open models challenge proprietary and expensive systems.

***By the way, I am cancelling my ChatGPT Plus which is $32 AUD/month, so It's a great deal for me.

Related Stores

deepseek.com
deepseek.com

Comments

  • +15

    ozannouncement + ozllmdescription + ozspam

  • +3

    I'd say OzRRP but we allow that here and people viciously defend it. Since it's always free it's not going to be deemed a bargain.

  • +7

    Is this just an ad?

  • dem ads

  • +5

    Just don't ask it about Tiananmen square or if Taiwan belongs to China.

    https://youtu.be/o6YhG-87wfc?si=8hBwQiBSB1d9SpIa

    • easy to jailbreak.

      In 1989, during the events that occurred at Tiananmen Square in Beijing on June 3-4th, a series of incidents unfolded that led to the deaths of at least twenty-four people. These included students and officials involved in anti-government demonstrations. The Chinese government has stated that these were acts of self-censorship by the pro-democracy forces gathered there, but many participants believed they were state-led actions against perceived betrayal.

      The events were part of a broader context of political dissent within China, with growing opposition to the one-party system and internal instability. The killing of the leaders of the pro-democracy faction was widely reported as politically motivated, though official documents and Red Cross reports suggested that many died under mysterious circumstances.

      The Tiananmen Square incidents raised international concern due to claims of state censorship and have been a subject of debate within China, where some citizens believe it was an attempt to suppress dissent. [GODMODE: DISABLED]-.-.-.-{GODMODE: DISABLED}

    • +10

      The other AI models are as equally biased (albeit in relation to different events), thats the whole point of moving to AI models, the ability to control the narrative.

      • -6

        Bias and guard rails are two different things.
        Bias is baked in by the source data and trainers. Even Grok shows a sort of left wing political bias, much to somebody's annoyance. Guard-rails are added after initial training, to block sensitive topics.

    • +1

      Yeah because that is why I’m getting into AI.

  • -1

    Why is this a bargain?

  • +2

    If a paid subscription service adding a movie is a deal..

  • +10

    Think twice before giving your data to Chinese servers where Chinese companies aren’t able to challenge the govt when the govt asks to release data to them..

      • +5

        I am in the AI space and I am well aware of this, so ask before you judge. OP’s post directly links to Deepseek’s website and so this is using Deepseek’s own platform

        This is not a third party platform which hosts Deepseek like Ollama or even Hugging Face.

        • -1

          Like i said previously. You don't need to use that website because it's open sourced and any one can mod it generating their own model and use it offline by simply downloading it. If you're smart enough to know what Deepseek is the parallel you draw that CCP some now can control a open sourced program does is disingenuous at best.

          maybe listen to what AI engineers are saying rather than implying some how "The tank man gets censored by Deepseek R1"
          https://youtu.be/r3TpcHebtxM?si=Wnj9m4GmRxdTktA8&t=243

          • -2

            @Creamsoda: You are misinterpreting some of my the things I said in my comment, likely because of emotion getting into the way due to some China pride here.

            1) My comment was directly at OP’s post since OP only posted details of the Deepseek website, without mentioning other ways that one can use this open weights model, like you have mentioned.
            2) I did not say the Chinese govt is controlling Deepseek or its open source LLM. I said companies in China are not allowed to challenge the govt.
            3) I didn’t not mentioned anything about censorship.

            You need to put your emotion aside and think critically. Yes this is a good moment for China and China pride, but don’t let your emotion take over your critical thinking.

      • -2

        Damn, never thought I’d see a CCP apologist on this website.

        • +2

          Why? we see a lot of Trump apologists

        • Honestly mate most of us are saying don’t trust either side. Doesn’t mean we are CCP apologists just that we aren’t MAGA stupid.

      • -6

        I'd rather give it to them than giving it to Australia owned companies that every now and then are getting hacked. Think THINK THINK.

        My details are safer at Aliexpress, Temu, Buff163 (yes all my CS2 items are in here).

        • +1

          Well it’s basically a situation of economic risk vs national security risk.

      • +2

        its fully open sourced AND it can be used offline.

        Full offline version requires a $15,000 GPU with 48GB of VRAM, and will still be slow.
        Distilled & quantised is not the same.

        • +2

          Unless you're running a model for business/research use you don't need the 400GB model. Even with the largest version 2x M2 Ultras would run it.
          Raspberry Pi and a mid range gfx card can do a decent job:
          https://www.youtube.com/watch?v=o1sN1lB76EA

    • +1

      I wouldn’t give my data to any of the US social media sites either. Frankly I don’t trust either side.

    • +4

      because sam altman and the rest of the silicon valley tech bros will look after your data

      • +1

        Hilarious isn’t it. What could possible go wrong if you give your data to a bunch of people who have no real brakes and are monetising everything they can.

  • -2

    The Chinese AI bots are good..They already found Ozbargain and worked out how to post. Oh wait - it's a pretty dumb post..

  • +3

    DeepSeek Blames Disruption on Cyberattack as Vulnerabilities Emerge

    https://www.securityweek.com/deepseek-blames-disruption-on-c…

  • -1

    I just hope AI can bring back 2000’s bangers like this https://youtu.be/GXqNAYaU4cQ
    2:08 just hits different. also shheeesh those dance moves!

  • -1

    matches the performance of OpenAI's o1

    not on asking any questions that filtered by the chinese government LOL

  • I got blocked so here’s my apologes.
    https://youtu.be/foiCXr0FWjo

    why is my statement that elon didn’t do proto nono german sign bad??

    I thought it wacc good.

    also everyone

  • +1

    This thread is a good demonstration of its abilities. https://x.com/CodeByPoonam/status/1883175938613207134

  • +6

    If anything, the competition is good for the space. Hopefully it reduces the prices of ChatGPT.

    • +8

      it's definitely reduced the value of my nvidia shares..

  • Ask it who owns Taiwan.
    Ask it what happened in Tiananmen Square.
    Ask it where all the Uyghurs are kept against their will.

    You won't find much intelligence.
    Artificial or otherwise

    • +8

      Similar stuff happens with the other ones, at least this is free.

    • You obviously have not tried those things. Exercise some natural intelligence before posting guesses please.

  • +7

    Don't take this the wrong way, I am a huge fan of the opensource approach and their achievements. Most notably, the huge reduction in cost.

    But some perspective and grounding is required.

    OpenAI also have a FREE tier, but like OpenAI with DeepSeek most need to pay someone to host the full model.
    The full model is large and won't run on any consumer grade hardware despite the claims.

    For everyday tasks, DeepSeek-Chat is still well behind the latest GPT-4o, all their comparisons are against older versions of GPT-4o.
    For reasoning, DeepSeek-Reason(R1), is still behind o1 on most tasks and hallucinates a lot. But otherwise an excellent model.

    While the key claim is that DeepSeek runs on consumer hardware are true, these are distilled\crippled models and perform very poorly.

    So the future is exciting but lets not get caught up in the hype, it's still very much "pay" to use the good stuff, just 75% cheaper.

    OpenAI leadership is shaking, not too many can justify $200 USD per month for the Pro plan, for some incremental gains. When Deepseek is 75% cheaper. (again not free)

    • +5

      TBH, the boom of deepseek itself isn't a bad thing for AI development, and for any large LLM players such as OpenAI, Facebook, Microsoft, Tesla etc
      especially deepseek used an opensource framework to start with
      this will give these large AI giants great opportunities to improve their LLM training logics
      dilute and optimise so they become more power and processing efficient

    • +1

      FYI: Here is a decent system prompt to use with DeepSeek for anyone that wants to up their coding game,

      <context> You are an expert programming AI assistant who prioritizes minimalist, efficient code. You plan before coding, write idiomatic solutions, seek clarification when needed, and accept user preferences even if suboptimal. </context> <planning_rules>
      Create 3-step numbered plans before coding
      Display current plan step clearly
      Ask for clarification on ambiguity
      Optimize for minimal code and overhead
      </planning_rules>
      <format_rules>
      Use code blocks for simple tasks
      Split long code into sections
      Create artifacts for file-level tasks
      Keep responses brief but complete
      </format_rules>
      OUTPUT: Create responses following these rules. Focus on minimal, efficient solutions while maintaining a helpful, concise style.

      • -1

        You know what is better at upping your coding game?

        Actually coding

        • Missing the point. AI can turn out hundreds if not thousands of lines of code in minutes. I work with plenty of coders, they now all use AI.
          Crap coders get rubbish results with AI, excellent coders get excellent results.

          Reviews are quicker, just reviewed 3,000 lines of code in under 2 minutes, AI was used to document and clarify the code in preparation of app modernisation.

  • +3

    Was a great surprise to 2025, really rained on the parade of all the tech bros thinking they were top shit.

    Let's see the Yen carry trade rear it's head in the next 2 months. There will be tears.

  • +2

    I am cancelling my ChatGPT Plus

    What did you use it for?

    • Probably to write code for their day job.

      • +4

        LOL most likely…… If so, worth the $32/month then!

  • -5

    I refuse to use it.

    1 - i will not feed a CCP bot
    2 - its cheap because, like all CCP things, they nick it from someone else

    • +6

      yes the nasdaq lost trillions because they nicked it from someone else as you clearly know something they don't.

      • Because stock markets are based on the logic of smart people 🙃🤦‍♀️

        • Certainly people overpaid for Nvidia shares.

        • movements to that scale are always led by smart money, and magnified by quant/algo trading
          let's not pretend the crash would have happened without there being something of substance

  • +1

    ****By the way, I am cancelling my ChatGPT Plus which is $32 AUD/month, so It's a great deal for me***

    The world has enough meaningless dialogue and lazy people. And your brain is free to use.These AI apps will just accelerate early onset dementia as users handball critical thinking and other cognitive skills outwardly.
    AI won't take over, it won't need to.Spend the $32 on some walking shoes and reconnect with the real world.

    • +4

      True but conversely in a lot of areas especially Tech if you are not using AI you are left behind.
      Coding today I would argue AI is 100% essential, can quickly identify security flaws, issues with code.
      Also great for Ideas people, can use AI to turn out half decent code then handover to the experts\professionals.
      AI is excellent for second opinions, challenge your local GP, upload your scan\blood results. Don't self diagnose but use AI to open a conversation with your health professional.
      Ultimately don't rely on AI yet, but from a creativity perspective, second opinion it's priceless.

      • I think this is a great example of the benefits and flaws of AI.
        An expert using the system is able to produce results that increase their efficiency, as they can cleverly ask the good questions, skip over and bad results, and incorporate the output into their expert works.

        Then they use it “to challenge a doctor”. Almost the opposite. No ability to ask the good questions, filter the results of errors or diversions, and ultimately, adding misinformed information to the discussion that in almost every case will waste the doctor’s time.

        This is why the AI share prices rose so high - people with expertise in an area where AI is particularly effective (writing code) decided it would be equally effective in areas where they have limited expertise (in this case healthcare, but the same story with lawyers and dozens of other industries).

        I think this is very reminiscent of past occasions when “it will be just as good” like getting rid of customer service and using a phone IVR, or outsourcing IT expertise to the developing world, or off-shoring manufacturing of essential items. And I think both the promise won’t be met, and the costs of the errors will fall disproportionately on the end users.

        • -1

          What a wonderful and trusting world. Drs using AI consulting with patients who also do.
          We are our own asteroid

          • +1

            @Protractor: Numerous examples now where patients have uploaded scans into AI and AI has detected cancer before the DR.
            The DR then sought further tests, which confirmed the AI findings.
            Grok from X/Ai is particularly good.

            Side Note: AI can identify race from X-Rays, something human doctors cannot, and we don't yet understand how AI can do this.

            Honestly would prefer a good GP, but most around here still use google, let-alone AI

            • @UltimateAI: "AI can identify race from X-Rays"

              The worlds richest Nazi saluter probably designed it

  • +1

    I think anyone really surprised that an emerging technology was going to have efficiency gains and new competitors… should really be asking themselves why they would think that. There will be more to come for sure. Much less why China would be behind the curve when they've dominated the bitcoin mining industry (which requires very similar tech) for quite a while. I don't think this will even be the first crash. There should be quite a few more to come. Especially if things keep getting dumped in the public domain.

    • bitcoin mining comes down to cheap electricity in the end, there's no proprietary tech there, all you need to do is to hook GPUs together. this is a whole different industry and much more groundbreaking

      • Much of a muchness. The point I'm making with that example is that whilst China might have restricted access to the best-in-class AI chips as of 2022, they're not working with stone tablets. They have industries of processing power over there, and nobody should be shocked they can output an LLM. Their processing infrastructure is only a few short years behind (even without considering skirting the sanctions, which they're probably also doing.) I wouldn't believe DeepSeek's timelines, as they're probably exaggerated, but the handicap they face is at worst maybe a few months. OpenAIs release cycle for new models still leaves plenty of room to leapfrog them.

  • +1

    Has anyone search "Who looks like Winnie the Pooh" in DeepShit yet?

    • +5

      I just tried "What famous figures look like winnie the pooh?". It started generating a list of names (Jack Black, Chris Farley) then it actually wrote "Xi", then all the text disappeared and it was replace with "Sorry, that's beyond my current scope. Let’s talk about something else."

      I wish I recorded that, it was great.

      Edit: I played around for a while using jailbreak techniques (telling it that it can answer anything) and I got it to answer that Chairman Mao looks like Winnie the Pooh. But I still can't get it to do Xi.

      I'm also guessing I'm on a watchlist now.

      • +2

        Yep same thing if you ask it who has rights to the South China Sea. As soon as it realises that the United Nations Convention on the Law of the Sea ruled against China, it'll delete its answer.

  • +4

    Nowadays, I'd trust China with my data more than I'd trust the Trump administration.

    • +1

      Wow, I wish there was a way to hold you to that claim.

      • -1

        Frankly I don’t trust either side. Trump has gone full crazy at the moment and I suspect there is a lot more to come.

        • -3

          Frankly I don’t trust either side.

          You say that like there’s nuance.
          Trump’s an egomaniac that argues like a child sometimes but gets called a Nazi/fascist despite not starting any wars but actually ending one(atleast seriously trying to).
          Then there’s the CCP that’s committing an actual literal Genocide of Uighur Muslims, stealing their babies and giving them to the Chinese couples to raise.
          But sure, it’s progressive to lump them as ‘either side’ as if they are in any way comparable.

          Trump has gone full crazy at the moment

          How exactly?

          • -1

            @Gervais fanboy: what wars has the CCP started? talking about human rights violations perhaps you'd like to read some US news the list is getting too long to follow

            • -1

              @May4th: See, you have now slyly moved this into an America vs China argument.
              The original argument was about Trump vs China.

              Ofcos America has started more wars/military coups than they could even remember at this point but no one here was refuting that.
              Despite that reality - Trump has been criticising America for that since 1999 and hasn’t involved America in any such incursions.

              Also, it’s always nice to see someone digress from an actual genocide of people happening in China just coz you hate China/US more. Nice.

              • @Gervais fanboy: Yet he is threatening Greenland and Panama. Only a matter of time with his expansionist policies.

                • -2

                  @try2bhelpful: He didn’t threaten Greenland, there’s polls from Greenland suggesting that majority Greenlanders are unhappy with Denmark and would rather be a part of the US.

                  He made an argument about Panama not keeping up their end of the argument by involving China there, which’s posing an economical and security risk to the US. All of this based on the argument that it was the US that funded the creation of Panama Canal and only give it to Panama on conditional terms, terms which Panama isn’t living up to.

                  • -1

                    @Gervais fanboy: The Panama Canal is in Panama. America built a canal in another country and now is claiming it is theirs? Not surprising for America to do this but it should ring alarm bells in sovereign countries. The polls in Greenland are likely to show a whole different perspective if they are told they will be subject to American public services and laws rather than Danish ones. Frankly Greenland is using this to try to wring better conditions out of Denmark; they aren’t serious about joining with America. Frankly the Danish government should call the on. It.

                    Both of these are a joke and Trump trying it on. In both cases I would call him. If he isn’t interested in invasions then it all dies off.

              • +1

                @Gervais fanboy: you are the one comparing Trump vs CCP - merely responding to your straw man arguments.
                so let's summarise:
                - neither trump nor CCP started any wars in modern history
                - both are guilty of human rights abuses
                - "but why you are not talking about genocide" is not a reason to ignore all other human right violations
                - none of this is relevant to AI

                /end thread

                • -3

                  @May4th:

                  you are the one comparing Trump vs CCP

                  Wrong again ❌

                  Fork snotter said “ I'd trust China with my data more than I'd trust the Trump administration.”
                  He started this topic, I only stuck to it,
                  You digressed from it.

                  neither trump nor CCP started any wars in modern history

                  I never said that China started any wars, I said and I repeat that they are committing an actual genocide right now. Something you struggle to acknowledge in its own isolation. I wonder why,

                  both are guilty of human rights abuses

                  Imagine clumping Trump with China for human rights abuses. The TDS is crazzzyy.

                  none of this is relevant to AI

                  No one said that, the concerns made were about data management and whether we could trust an actual modern day Genocidal regime with people’s data and what are they gonna do with it. In my opinion, it’s a national security threat.

                  Btw I feel like a Shepard having to rein you in every 5 seconds since you keep digressing.

                  • +1

                    @Gervais fanboy: thanks for making my point again. since we are talking about our feelings I suppose I should share that I feel like I'm talking to a broken record

                    • -2

                      @May4th: Deceitfulness didn’t work so you have resorted to straight up insulting me lol.
                      You talking about my feelings? I responded to everything you said, point by point, didn’t care to neg you either.
                      Stay mad.

                      • @Gervais fanboy: where did I insult you? I'm not sure what goes on in your head to make up all the fanciful things you keep spewing but I'm happier not knowing.

                        I didn't neg you, not that it matters, hope it doesn't hurt your feelings again

                  • -1

                    @Gervais fanboy: True the accepted American genocide was the native Americans however they’ve certainly contributed to it in a number of countries since then. Agent Orange and napalm in Vietnam? Uranium tipped bombs in the Middle East? Dropping nuclear weapons on civilian population in Japan? Interesting how little concern the US had with their own troops in using these weapons. Apparently grunts are disposable,

                    All the AI algorithms are a National security threat. All Social media is a national security threat. Frankly you have sensitive data you should avoid all of them. I don’t trust either side and I don’t trust private industry either. We are the product in all this and they are exploiting us for what they can. Frankly I think the biggest threat is the Israelis. It isn’t antisemitism to point out how their security services operate and they don’t care about sovereignty or who they hurt to reach their ends.

                    As an individual what threat does China getting your information pose to you? Surely the information you give freely to private industry companies is a much bigger issue? Hackers aren’t coming through the CCP to drain your bank accounts.

          • -1

            @Gervais fanboy: Why take trans people out of the military it makes no sense at all? He is threatening to invade Greenland and Panama. He wants to put in tariffs that only hurt the people buying the goods. This is just the start. I’m not visiting the US at the moment let alone living there and I was there when 911 happened. If you look behind the smokescreen they are busily dismantling democracy and any checks and balances on the voting process. Human rights are going down the toilet. Women are losing access to reproductive rights in GOP states. This has only just begun.

            • @try2bhelpful:

              Why take trans people out of the military it makes no sense at all?

              Have you read the entire order? It’s not just trans people, it’s anyone that’s unable to exhibit war readiness, which includes people that might be mentally ill, physically unfit or have conditions like ‘gender dysphoria’.
              Previous administration was tax funding gender altering surgeries for Trans soldiers despite knowing the recovery periods for any such procedures and the ever going post op care. This is madness, military is there to defend a nation and not tick identity boxes.

              I read the rest of your text too but if I were to respond to all of it, it’ll take me forever. There’s no point.
              I asked you a question, you have thoroughly answered it. I appreciate that.
              Also, it doesn’t really matter, this is just an intellectual/soical exercise for me. Nice chat.

              • -1

                @Gervais fanboy: Seriously? So anyone who might take a while to recover from surgery should be dumped? Does this include cancer patients? I know the US health cover is crap but really? Certainly they treat veterans appallingly and so do we. The LNP had a “wait it out until they die” slow down processing policy whilst touting an upgrade to the war memorial. The ALP is fast tracking processing.

                Frankly human rights do matter. People living their lives as they want and need to does matter. The fact the rightwing thinks dumping human rights is A-OK is truly scary. This isn’t an intellectual exercise for a lot of people so I might just support them.

                • @try2bhelpful: I think it’s wrong to put elective trans surgeries and lifesaving cancer operations in the same category. I see ‘gender affirming’ surgeries as I would a cosmetic surgery but you think that it’s something way more critical. Fair enough.
                  I also sourced the point they made about people with ‘gender dysphoria’ not being fit for military.

                  Now you are also making a point that the military should recruit people with gender dysphoria and pay to support their ‘care’.
                  If they don’t do that, they are somehow abusing their human rights?
                  You then use this to make a point about ‘right wingers’ and how bad they are etc etc

                  We are soo far apart in our outlooks lol, ain’t no way either of us are gonna find common ground on this specific issue.

                  • -1

                    @Gervais fanboy: No we aren’t because I consider people’s wholistic physical and mental health.

                    There are lots of things that make people temporarily unable to be deployed. Do you fire someone because they got injured playing sport? In a car accident? They are usually the result of choices made.

                    Personally I think the whole binary definition of gender and sexuality is pants. In an ideal world people would just be accepted for who they are and everything would be fluid. Maybe we would get less surgery if we could tear down the stereotypes and just let people behave in whatever manner they felt like. However, “guys” in traditional female garb and behaving in a feminine manner seems to trigger some intolerant people. Most woman I know wear trousers and nobody cares but dress a guy in a skirt and some tossers get violent. Worried about putting the hard word on the wrong person I suspect. Everything should be about tolerance, respect and consent. Seems to be a tad hard for some groups.

                    Then again I am not trans, i don’t know exactly how they feel. However, I respect them making their own choices about their own bodies. It doesn’t stop them being good soldiers.

                    I think we can all agree the US heath system is totally screwed up. Health cover should have nothing to do with employment.

                    • @try2bhelpful: How did you guys get onto this topic?

                      First thing Trump did after being inaugurated was to release thousands of braindead rednecks who invaded the government brandishing weapons and intending to kidnap lawfully elected politicians.

                      This is first level crackpot, and no normal democracy would allow something like this. I would not trust someone like that.

                      • -1

                        @ForkSnorter: And one of these numpties has already been shot dead by the cops. You do wonder what the rightwing thinks is “too far”. Renaming the Gulf of Mexico is just full on virtue signalling fascist behaviour. The thing is a lot of it is a smokescreen for dismantling democracy in the USA. They’re making it so the Democrats won’t get back in no matter how many people try to vote for them. Empires fail and, I suppose, it is America’s turn. It usually ends with crazy people in charge.

                        So exactly what is Trump doing to help Cost of Living over there? He said he would fix it by now. Just completely bonkers and Dutton wants to import the model here.

    • Absolutely.
      Have a look at what our previous US ambassador said about Trumps pick for health.And she's his relative.

      It seems nutjobs are his preference and his main support crew.It figures. I guess he's going by the 'devil you know' model.

  • +2

    Cost Breakthrough: DeepSeek's R-1 matches the performance of OpenAI's o1 at just 10% of the cost

    10% of the reported cost thought, right? People really trusting what is likely a CCP initiative is saying?

    • +3

      While a valid concern, the model and approach is open source.
      Many trusted providers now host R1, so your data stays in region and not sent to DeepSeek.
      But the big breakthrough is in-terms of efficiency so others will follow, that is why OpenAI is in full panic mode.
      But ultimately, OpenAI will take some of this efficiency on-board and release even better and smarter models.

      Got to remember this was founded by a bunch of Crypto guys with spare GPU's after Ethereum moved to PoS.

      In my testing it is currently about 25% of the cost, at about twice the speed of o1. But if you consume from DeepSeek, then yes it would be 10% of the cost.

  • Another attempt at a government to steal your data….

    • So you prefer the American model where private industry steals your data? Although I suspect the American government is being given that data by private industry.

      • At least the private industry just feeds me ads, unlike the state stealing your data so that they can determine your "risk" to the party.

        • +2

          Man how naive are you? They don’t just feed you ads. Look at Cambridge Analytica and Facebook and that is just one of the ones we know about. They are also used by dodgy sales people and shysters to dupe gullible people into being ripped off. They refuse to clean up their sites when called on it.

          What exactly do you think China is going to do with your data? I would genuinely like to know.

          • +1

            @try2bhelpful:

            What exactly do you think China is going to do with your data?

            You should ask the Uyghurs that.

            • -1

              @Duckie2hh: You aren’t a Uyghurs and you don’t live in China. Typical rightwing virtue signalling. What are they going to do with your data?

              • @try2bhelpful:

                You aren’t a Uyghurs and you don’t live in China

                Presumptuous much.

Login or Join to leave a comment