Censoring is model dependent so you can select one of the models without the guardrails.

It would be so cool if we could generate the entire amount of future energy usage all at once and just store it. May be we could launch the energy store in space and place it so that, let’s say it takes about 8 minutes for the energy to reach us.

Apple fucked up no doubt. Given how hard they pushed AI as a key feature of IPhone 16 I wouldn’t be surprised if they get a class action lawsuit for this.

But it’s also interesting to read a few things from the article that makes me hopeful for when Apple finally releases the features:

1. Let’s be honest, AI by Google, MS is shit right now. They are claiming the same promises which most of the time don’t work, but Apple chose to delay release until they could get better consistency.
2. The executives are taking personal responsibility? I hope that’s the case and no developers are thrown under the bus for this. I’ve rarely seen an article mention personal executive responsibility from a tech firm for delays and qa issues.
3. I hope marketing gets reigned in so they won’t push other unready features the next few years.
4. I hope Apple releases some open source AI tooling to re-gain good will. Would love to see some more competition in the AI space.

1. Rampant unchecked capitalism of recent decades has created large wealth disparities akin to the earlier decades of the last century. It is no longer possible for one person in a household with a regular job to support a modest lifestyle for their family. All benefits especially medical for the whole family, being completely intertwined with the current job reduces mobility and further feeds into the wealth gap by keeping wages low. It’s easier to blame the powerless for this state of affairs than the powerful because the powerless cannot object.
2. The fear of the other has been accentuated by media and misinformation. Targeted algorithms feeding most of the information that is consumed has created echo chambers that reinforce existing beliefs and fears. The propaganda state has never had it easier.
3. The large military and police has given never before control to the state about what is allowed to be protested. Combined with the day to day struggles, it’s extremely hard to come together for what is right. The ruling class is able to maintain the fine balance between absolute misery and general dissatisfaction that it is still better to struggle through a thankless job than to say fuck it. Failures of recent large uprisings like Middle East and Hong Kong have reinforced the futility of standing up against the rulers.
4. Evil has many heads and there’s always one head that you can find alignment with. It could be the deregulation of businesses, lower taxes, anti abortion, racism, but as long as there’s one thing you can align on, the general sense of powerlessness makes it easier to overlook the other heads.
5. The line between evil and good has never been murkier, especially with globalization. If you focus on the betterment of your community, it would be considered good, but what if it leads to suffering of others outside the community. Is it also evil? What is community - is it the people in your neighborhood, your religion, your country, fellow business owners? The fuzzier these lines are, the harder it is to untangle them.

Depending on the state you live in, and whether the lease specifically mentioned taking special care of the hardwood floors, this could very well be considered regular wear and tear. There does not appear to be intentional damage here.

The landlord will also need to provide documentation prior to getting work done if they want you to cover the bill, at which point you’ll have the option to contest it.

Check out tenant rights for your state to verify. Hopefully, you’ll also be more careful with hardwood floors in the future. Couches on wheels are no-bueno.

This looks like a marketing email because you agreed to share information with Shop Pay as the first line says. It may have nothing to do with your specific order shipment.

Not to say it’s not annoying to see another “please please download our app” message, but the pitchforks aren’t necessary yet.

• \cdot for vectors and in some cases, scalar variables, e.g., x \cdot y
• \times for numbers, e.g., 2 \times 8
• nothing for combination and scalars, e.g., 2x or xy

Late 90s to 2000s was the decade of internet glory. Then social media and big tech took over. Now with personalized feeds and searches, along with conflict promoting engagement metrics, many people spend their time within echo chambers and those chambers keep getting more partisan. On top of that, rampant misinformation has made it all the more difficult to separate fact from fiction.

The bill excludes holding responsible creators of open source models for damages from forked models that have been significantly altered.

The criticism from large AI companies to this bill sounds a lot like the pushbacks from auto manufacturers from adding safety features like seatbelts, airbags, and crumple zones. Just because someone else used a model for nefarious purposes doesn’t absolve the model creator from their responsibility to minimize that potential. We already do this for a lot of other industries like cars, guns, and tobacco - minimize the potential of harm despite individual actions causing the harm and not the company directly.

I have been following Andrew Ng for a long time and I admire his technical expertise. But his political philosophy around ML and AI has always focused on self regulation, which we have seen fail in countless industries.

The bill specifically mentions that creators of open source models that have been altered and fine tuned will not be held liable for damages from the altered models. It also only applies to models that cost more than $100M to train. So if you have that much money for training models, it’s very reasonable to expect that you spend some portion of it to ensure that the models do not cause very large damages to society.

So companies hosting their own models, like openAI and Anthropic, should definitely be responsible for adding safety guardrails around the use of their models for nefarious purposes - at least those causing loss of life. The bill mentions that it would only apply to very large damages (such as, exceeding $500M), so one person finding out a loophole isn’t going to trigger the bill. But if the companies fail to close these loopholes despite millions of people (or a few people millions of times) exploiting them, then that’s definitely on the company.

As a developer of AI models and applications, I support the bill and I’m glad to see lawmakers willing to get ahead of technology instead of waiting for something bad to happen and then trying to catch up like for social media.

If I understand correctly from the article, you have to enter ‘OOBE\BYPASSNRO’ in command prompt during installation to prevent it from asking to connect to internet. If that’s the only way to set up a local account, that’s hardly an accessible option.

Wouldn’t it be possible to buy a new PC, open the box, and return it right after because you cannot set it up without internet?

If enough people do it, may be PC manufacturers will force Microsoft to add offline setups.

One of the funniest things about most of these companies enforcing RTO is that their “on-site interviews” are still virtual. So you believe being in-person is more effective except when it comes to paying for travel expenses for interviewees.

Just shows the massive hypocrisy behind these RTO mandates.

I reached out to Roku support regarding this. The rep told me “why are you complaining. You are the only one.” He then disconnected the chat. I’ve reached out to my state’s AG to report this. No action so far but waiting. If there are enough complaints, that might help move the needle.

What Roku is doing should be completely illegal - bricking the product after purchasing it for full price if you don’t agree to waiving your rights.