#AI Project Estimation

[lostconsultants]

Embracing AI-Driven Estimations for a Smarter Future

Embracing AI-driven estimations for a smarter future is no longer a distant vision—it is a necessary step in modern project management. In previous discussions, we explored the foundation of AI-driven estimations in Rethinking Estimations in the Age of AI and how AI is actively revolutionising estimation practices in Revolutionising Estimations with AI: Smarter, Faster, and More Reliable…

[constructionsoftware]

Discover how construction project management software optimizes project planning, scheduling, and execution. This powerful tool improves collaboration, ensures timely completion, and reduces costs, providing construction professionals with the resources needed for successful project delivery.

[derinthescarletpescatarian]

what’s the story about the generative power model and water consumption? /gen

There's this myth going around about generative AI consuming truly ridiculous amount of power and water. You'll see people say shit like "generating one image is like just pouring a whole cup of water out into the Sahara!" and bullshit like that, and it's just... not true. The actual truth is that supercomputers, which do a lot of stuff, use a lot of power, and at one point someone released an estimate of how much power some supercomputers were using and people went "oh, that supercomputer must only do AI! All generative AI uses this much power!" and then just... made shit up re: how making an image sucks up a huge chunk of the power grid or something. Which makes no sense because I'm given to understand that many of these models can run on your home computer. (I don't use them so I don't know the details, but I'm told by users that you can download them and generate images locally.) Using these models uses far less power than, say, online gaming. Or using Tumblr. But nobody ever talks about how evil those things are because of their power generation. I wonder why.

To be clear, I don't like generative AI. I'm sure it's got uses in research and stuff but on the consumer side, every effect I've seen of it is bad. Its implementation in products that I use has always made those products worse. The books it writes and flood the market with are incoherent nonsense at best and dangerous at worst (let's not forget that mushroom foraging guide). It's turned the usability of search engines from "rapidly declining, but still usable if you can get past the ads" into "almost one hundred per cent useless now, actually not worth the effort to de-bullshittify your search results", especially if you're looking for images. It's a tool for doing bullshit that people were already doing much easier and faster, thus massively increasing the amount of bullshit. The only consumer-useful uses I've seen of it as a consumer are niche art projects, usually projects that explore the limits of the tool itself like that one poetry book or the Infinite Art Machine; overall I'd say its impact at the Casual Random Person (me) level has been overwhelmingly negative. Also, the fact that so much AI turns out to be underpaid people in a warehouse in some country with no minimum wage and terrible labour protections is... not great. And the fact that it's often used as an excuse to try to find ways to underpay professionals ("you don't have to write it, just clean up what the AI came up with!") is also not great.

But there are real labour and product quality concerns with generative AI, and there's hysterical bullshit. And the whole "AI is magically destroying the planet via climate change but my four hour twitch streaming sesh isn't" thing is hysterical bullshit. The instant I see somebody make this stupid claim I put them in the same mental bucket as somebody complaining about AI not being "real art" -- a hatemobber hopping on the hype train of a new thing to hate and feel like an enlightened activist about when they haven't bothered to learn a fucking thing about the issue. And I just count my blessings that they fell in with this group instead of becoming a flat earther or something.

[batboyblog]

Things Biden and the Democrats did, this week #16

April 26-May 3 2024

President Biden announced $3 billion to help replace lead pipes in the drinking water system. Millions of Americans get their drinking water through lead pipes, which are toxic, no level of lead exposure is safe. This problem disproportionately affects people of color and low income communities. This first investment of a planned $15 billion will replace 1.7 million lead pipe lines. The Biden Administration plans to replace all lead pipes in the country by the end of the decade.

President Biden canceled the student debt of 317,000 former students of a fraudulent for-profit college system. The Art Institutes was a for-profit system of dozens of schools offering degrees in video-game design and other arts. After years of legal troubles around misleading students and falsifying data the last AI schools closed abruptly without warning in September last year. This adds to the $29 billion in debt for 1.7 borrowers who wee mislead and defrauded by their schools which the Biden Administration has done, and a total debt relief for 4.6 million borrowers so far under Biden.

President Biden expanded two California national monuments protecting thousands of acres of land. The two national monuments are the San Gabriel Mountains National Monument and the Berryessa Snow Mountain National Monument, which are being expanded by 120,000 acres. The new protections cover lands of cultural and religious importance to a number of California based native communities. This expansion was first proposed by then Senator Kamala Harris in 2018 as part of a wide ranging plan to expand and protect public land in California. This expansion is part of the Administration's goals to protect, conserve, and restore at least 30 percent of U.S. lands and waters by 2030.

The Department of Transportation announced new rules that will require car manufacturers to install automatic braking systems in new cars. Starting in 2029 all new cars will be required to have systems to detect pedestrians and automatically apply the breaks in an emergency. The National Highway Traffic Safety Administration projects this new rule will save 360 lives every year and prevent at least 24,000 injuries annually.

The IRS announced plans to ramp up audits on the wealthiest Americans. The IRS plans on increasing its audit rate on taxpayers who make over $10 million a year. After decades of Republicans in Congress cutting IRS funding to protect wealthy tax cheats the Biden Administration passed $80 billion for tougher enforcement on the wealthy. The IRS has been able to collect just in one year $500 Million in undisputed but unpaid back taxes from wealthy households, and shows a rise of $31 billion from audits in the 2023 tax year. The IRS also announced its free direct file pilot program was a smashing success. The program allowed tax payers across 12 states to file directly for free with the IRS over the internet. The IRS announced that 140,000 tax payers were able to use it over their target of 100,000, they estimated it saved $5.6 million in tax prep fees, over 90% of users were happy with the webpage and reported it quicker and easier than companies like H&R Block. the IRS plans to bring direct file nationwide next year.

The Department of Interior announced plans for new off shore wind power. The two new sites, off the coast of Oregon and in the Gulf of Maine, would together generate 18 gigawatts of totally clean energy, enough to power 6 million homes.

The Biden Administration announced new rules to finally allow DACA recipients to be covered by Obamacare. Deferred Action for Childhood Arrivals (DACA) is an Obama era policy that allows people brought to the United States as children without legal status to remain and to legally work. However for years DACA recipients have not been able to get health coverage through the Obamacare Health Care Marketplace. This rule change will bring health coverage to at least 100,000 uninsured people.

The Department of Health and Human Services finalized rules that require LGBTQ+ and Intersex minors in the foster care system be placed in supportive and affirming homes.

The Senate confirmed Georgia Alexakis to a life time federal judgeship in Illinois. This brings the total number of federal judges appointed by President Biden to 194. For the first time in history the majority of a President's nominees to the federal bench have not been white men.

[reasonsforhope]

"The first satellite in a constellation designed specifically to locate wildfires early and precisely anywhere on the planet has now reached Earth's orbit, and it could forever change how we tackle unplanned infernos.

The FireSat constellation, which will consist of more than 50 satellites when it goes live, is the first of its kind that's purpose-built to detect and track fires. It's an initiative launched by nonprofit Earth Fire Alliance, which includes Google and Silicon Valley-based space services startup Muon Space as partners, among others.

According to Google, current satellite systems rely on low-resolution imagery and cover a particular area only once every 12 hours to spot significantly large wildfires spanning a couple of acres. FireSat, on the other hand, will be able to detect wildfires as small as 270 sq ft (25 sq m) – the size of a classroom – and deliver high-resolution visual updates every 20 minutes.

The FireSat project has only been in the works for less than a year and a half. The satellites are fitted with custom six-band multispectral infrared cameras, designed to capture imagery suitable for machine learning algorithms to accurately identify wildfires – differentiating them from misleading objects like smokestacks.

These algorithms look at an image from a particular location, and compare it with the last 1,000 times it was captured by the satellite's camera to determine if what it's seeing is indeed a wildfire. AI technology in the FireSat system also helps predict how a fire might spread; that can help firefighters make better decisions about how to control the flames safely and effectively.

This could go a long way towards preventing the immense destruction of forest habitats and urban areas, and the displacement of residents caused by wildfires each year. For reference, the deadly wildfires that raged across Los Angeles in January were estimated to have cuased more than $250 billion in damages.

Muon is currently developing three more satellites, which are set to launch next year. The entire constellation should be in orbit by 2030.

The FireSat effort isn't the only project to watch for wildfires from orbit. OroraTech launched its first wildfire-detection satellite – FOREST-1 – in 2022, followed by one more in 2023 and another earlier this year. The company tells us that another eight are due to go up toward the end of March."

-via March 18, 2025

[draconym]

nightshade is basically useless https://www.tumblr.com/billclintonsbeefarm/740236576484999168/even-if-you-dont-like-generative-models-this

I'm not a developer, but the creators of Nightshade do address some of this post's concerns in their FAQ. Obviously it's not a magic bullet to prevent AI image scraping, and obviously there's an arms race between AI developers and artists attempting to disrupt their data pools. But personally, I think it's an interesting project and is accessible to most people to try. Giving up on it at this stage seems really premature.

But if it's caption data that's truly valuable, Tumblr is an ... interesting ... place to be scraping it from. For one thing, users tend to get pretty creative with both image descriptions and tags. For another, I hope whichever bot scrapes my blog enjoys the many bird photos I have described as "Cheese." Genuinely curious if Tumblr data is actually valuable or if it's garbage.

That said, I find it pretty ironic that the OP of the post you linked seems to think nightshade and glaze specifically are an unreasonable waste of electricity. Both are software. Your personal computer's graphics card is doing the work, not an entire data center, so if your computer was going to be on anyway, the cost is a drop in the bucket compared to what AI generators are consuming.

Training a large language model like GPT-3, for example, is estimated to use just under 1,300 megawatt hours (MWh) of electricity; about as much power as consumed annually by 130 US homes. To put that in context, streaming an hour of Netflix requires around 0.8 kWh (0.0008 MWh) of electricity. That means you’d have to watch 1,625,000 hours to consume the same amount of power it takes to train GPT-3. (source)

So, no, I don't think Nightshade or Glaze are useless just because they aren't going to immediately topple every AI image generator. There's not really much downside for the artists interested in using them so I hope they continue development.

[probablyasocialecologist]

The almost overnight surge in electricity demand from data centers is now outstripping the available power supply in many parts of the world, according to interviews with data center operators, energy providers and tech executives. That dynamic is leading to years-long waits for businesses to access the grid as well as growing concerns of outages and price increases for those living in the densest data center markets. The dramatic increase in power demands from Silicon Valley’s growth-at-all-costs approach to AI also threatens to upend the energy transition plans of entire nations and the clean energy goals of trillion-dollar tech companies. In some countries, including Saudi Arabia, Ireland and Malaysia, the energy required to run all the data centers they plan to build at full capacity exceeds the available supply of renewable energy, according to a Bloomberg analysis of the latest available data. By one official estimate, Sweden could see power demand from data centers roughly double over the course of this decade — and then double again by 2040. In the UK, AI is expected to suck up 500% more energy over the next decade. And in the US, data centers are projected to use 8% of total power by 2030, up from 3% in 2022, according to Goldman Sachs, which described it as “the kind of electricity growth that hasn’t been seen in a generation.”

21 June 2024

[asestimationsconsultants]

The Digital Shift | How Tech is Redefining Industrial Estimating Service

Introduction: The Role of Technology in Modern Industrial Estimating The industrial sector, once reliant on traditional methods of estimating project costs, has entered a new era where technology is reshaping how projects are planned, managed, and executed. In particular, industrial estimating services are benefiting from the integration of cutting-edge technology that enhances accuracy, efficiency, and overall project outcomes. As industries adopt advanced digital tools, the way cost estimates are generated and managed has evolved significantly, streamlining workflows, improving decision-making, and reducing the risks associated with manual errors and outdated practices.

The Shift from Manual Estimating to Digital Tools For many years, industrial estimating was a manual process that involved spreadsheets, paper-based plans, and extensive calculations. While these methods worked, they were time-consuming and prone to human error. Today, however, digital tools such as Building Information Modeling (BIM), 3D modeling, cloud-based estimating software, and artificial intelligence (AI) are transforming how industrial projects are estimated.

These technologies provide estimators with more powerful, accurate, and efficient means of calculating project costs. For instance, BIM allows for the creation of detailed, 3D models of a project, enabling estimators to visualize the design and identify potential issues before construction begins. This can lead to more accurate estimates and fewer surprises during the construction phase.

Building Information Modeling (BIM): Revolutionizing the Estimating Process BIM has revolutionized the way construction projects are designed and estimated. In industrial projects, BIM provides a detailed, virtual representation of the entire project, including every structural element, material, and system. This digital model allows estimators to see the project in its entirety and calculate accurate costs based on real-time data.

BIM helps estimators identify potential conflicts or inefficiencies in the design early on, reducing the likelihood of costly mistakes. For example, by modeling the mechanical, electrical, and plumbing systems in 3D, estimators can spot clashes or design issues that may lead to delays or cost overruns. This proactive approach allows estimators to adjust the design or budget before construction begins, improving the overall accuracy of the estimate.

Moreover, BIM allows for seamless collaboration between different stakeholders, such as architects, engineers, and contractors. Estimators can share the digital model with the entire project team, ensuring that everyone is on the same page and working toward the same goal.

Cloud-Based Estimating Software: Collaboration and Accessibility Cloud-based estimating software has become a game-changer for industrial estimating services. Unlike traditional desktop-based software, cloud-based solutions allow estimators to access project data from anywhere, at any time. This enhances collaboration between team members, contractors, and clients, as everyone can work from the same platform and update the estimate in real time.

Cloud-based software also provides the flexibility to store and organize vast amounts of project data, including cost breakdowns, material lists, and labor schedules. This data is securely stored in the cloud, ensuring that project details are accessible and up-to-date, even as changes occur during the project's lifecycle.

Additionally, cloud-based software can integrate with other project management tools, such as scheduling and procurement software. This enables estimators to track changes in real-time and adjust the cost estimates accordingly. For example, if there is a delay in material delivery, the estimator can update the estimate immediately, helping to keep the project on track and prevent cost overruns.

Artificial Intelligence and Machine Learning: Predictive Estimating Artificial intelligence (AI) and machine learning (ML) are increasingly being used in industrial estimating services to improve the accuracy and efficiency of cost predictions. By analyzing vast amounts of historical data, AI and ML algorithms can identify patterns and trends that human estimators may overlook.

For example, AI can help predict material costs based on historical pricing data, accounting for factors like inflation and market fluctuations. This allows estimators to generate more accurate cost projections, even in volatile markets. AI can also help identify potential risks, such as supply chain disruptions, that may impact project timelines or costs.

Machine learning algorithms can continuously improve over time as they are exposed to more data, allowing estimators to refine their predictions and enhance their accuracy. The ability to leverage AI and ML in the estimating process helps reduce the likelihood of cost overruns and ensures that projects stay within budget.

Real-Time Data and Analytics: Making Informed Decisions In the past, estimators relied on static data, such as historical cost records and vendor quotes, to generate project estimates. However, with the rise of real-time data and analytics, industrial estimators now have access to more dynamic and up-to-date information that can impact their cost predictions.

For instance, cloud-based platforms and integrated project management systems allow estimators to access live pricing data, including current material costs, labor rates, and equipment rental fees. This real-time information helps estimators make more accurate predictions and adjust their estimates based on the latest market conditions.

Real-time data also allows estimators to track project progress more closely, monitoring any changes in scope or schedule that may affect costs. With up-to-date information at their fingertips, estimators can make informed decisions about how to adjust the budget or timeline, ensuring that the project stays on track and within budget.

Digital Integration with Supply Chain and Procurement Technology is also enhancing the way industrial estimating services interact with supply chain and procurement systems. With digital tools, estimators can easily integrate their cost estimates with procurement data, allowing them to track material availability, supplier lead times, and pricing fluctuations in real-time.

This integration helps prevent delays caused by supply chain disruptions and ensures that estimators can adjust their cost estimates as material prices fluctuate. For example, if a material becomes more expensive due to a global shortage, the estimator can immediately update the cost estimate and notify stakeholders of the potential impact on the project budget.

By streamlining the procurement process and connecting it directly to the estimating system, industrial estimators can avoid costly delays and keep the project moving forward.

The Future of Industrial Estimating: Virtual and Augmented Reality Looking ahead, virtual and augmented reality (VR and AR) are set to play an even more significant role in industrial estimating services. With VR and AR, estimators can create immersive, 3D simulations of the project, allowing them to visualize the construction process and identify potential issues before the project begins.

For example, by using AR glasses, estimators can overlay digital models onto the physical site, providing a real-time view of how the project will look once completed. This technology can help estimators identify issues related to space, design, or logistics that may not be apparent in traditional 2D drawings.

Incorporating VR and AR into the estimating process allows for better decision-making and a more accurate understanding of how the project will progress. As these technologies continue to evolve, industrial estimating services will become even more precise and efficient.

Conclusion: Embracing the Digital Shift for Better Industrial Estimating The digital shift is transforming the way industrial estimating services operate, bringing about improvements in accuracy, efficiency, and collaboration. From BIM and cloud-based software to AI and real-time data, technology is enabling industrial estimators to generate more precise cost estimates, reduce risks, and streamline the estimating process.

By embracing these digital tools, industrial estimating services can provide more reliable estimates that help projects stay within budget and on schedule. As technology continues to evolve, the potential for further improvements in estimating accuracy and efficiency is immense, and those who adopt these tools will be better equipped to navigate the complexities of modern industrial construction.

[lostconsultants]

Revolutionising Estimations with AI: Smarter, Faster, and More Reliable Predictions

Revolutionising estimations with AI transforms project planning by enhancing accuracy and reducing uncertainty. Unlike traditional methods prone to bias, AI-driven estimations leverage historical data and predictive analytics for more reliable forecasts.

Revolutionising estimations with AI is transforming how teams predict timelines, allocate resources, and improve project planning. Traditional estimation methods often rely on human intuition, which can introduce biases and inconsistencies. AI offers a data-driven approach that enhances accuracy, reduces uncertainty, and allows teams to focus on delivering value. If you’re interested in…

[emperor-kumquat]

Sincere suggestions for commissioned artists, from a client's perspective

I am an artist myself, but I don't do commissions.

However I do commission artists and pay them $1000s of dollars USD every year! My project Transformers: Mercy has included around 100 artists and I look for new ones all the time on Tumblr. After all, I just couldn't complete my project without some help. In the Transformers fandom, I guess I could be a top fan artist employer! Screw AI art, right?!

(Feel free to DM me if you would like to be commissioned by me or volunteer art for my big Transformers: Prime sequel game.)

When everything goes well, I am a repeat commissioner and you have me as a permanent client. But what might turn me or another client away from commissioning you initially or a second time?

If you...

... Ask for email-only communication/ have any barriers to direct and fast communication [X]

If we're on Tumblr, I would just like to message you all the details for my comm and get quick updates here. Emailing is not swift! Or sometimes, you have a setting where only blogs you follow can message you. I'll try to leave a comment or an ask, but I might not be seen and the opportunity passes. Consider turning DMs on if you are looking for commissions!

I am also not a huge fan of downloading a comm question sheet, filling it out, and sending it back. It is so much faster for me to just dump all the details of my request (and by now, I know how to be specific). If anything is unclear, I would like the artist to just reach out and ask.

2. ... Give no regular communication and updates [X]

If someone has paid you for a service, please respect that. Give an estimated time for completion and if something comes up, update the client. Clients get nervous waiting days or weeks without updates and they should not have to reach out to ask about what they paid for. Give progress updates too... sketch phase, line art, flats, rendering, what have you. If the client wants changes, it is less grief for everyone to make the change before too much effort goes into the comm. Please, reassure your clients!

3. ... Don't refer to given instructions and reference pictures [X]

I give very specific instructions and reference pictures for my artists. Sometimes, time passes. They forget what the instructions were and send me a work in progress that is not what was requested. It's frustrating when with a scroll up, they would have seen all the instructions again. Please look back to your instructions constantly throughout the process, not just once. Double-check for details as you work.

Client story: Sometimes it's as silly as me asking for a night scene and getting sent a day scene. Sometimes it's characters doing the wrong action. Sometimes it's the wrong angle from what I asked for. Sometimes it's forgetting that I wanted a character to be holding a certain item. All the instructions were given, and the chunk of text which describes the scene, but they were not carefully read.

4. ... State "no changes" or "limited number of changes" in your comm sheet [X]

This worries a client like me. What does that mean, I wonder? What if I perfectly expressed my request but the artist didn't get it right? Will they get upset if I request a small change? I have come to understand some people mean "no major changes from the original request", but it can sound like I don't have a say in the production phase and I could be stuck with something I don't like.

Client story: One time, an artist forgot a piece to a character's arm. It was a wing and blaster, clear on the reference sheet. I only noticed later after the comm was done, so that was on me. I reached out, hoping they could add in the missing piece even for a reasonable quick price because it took me so long to notice. They told me it would cost $40 USD to add in the piece of the arm. I declined. $40 USD, for a piece of an arm?? It was unreasonable. Because of that, I have not returned to this client and otherwise would have.

5. ... Ask for payment up front [X?]

It's a stand off! Because no one wants to be scammed. The artist is afraid of spending so much time and effort to not be paid; the client is afraid the artist is a fraud and will run away with the money.

Now I haven't been scammed yet and I have paid many artists up front many times. But I saw their blogs and they seem trustworthy. Whereas those bots on fanfiction.net, never trust them. But it is still uncomfortable as a client to pay upfront. Sometimes I pay an artist, then 6 months later, I get my art. If you need a lot of time to finish a piece, inform the client and consider getting your payment after. Care for your client and ease their worries. It isn't fun to pay someone then get radio silence for a quarter or half a year... imagine being in that position!

The best solution I've seen? The client pays full or half the payment after the rough sketch phase. The artist hasn't committed too much time yet and the client can see if the artist is capable or not of completing the request. I think it makes us all feel more at ease.

6. ... Are strict/inflexible with pricing [X]

An artist can decide their prices. This isn't about how to value art. But consider this story for nuance:

Client story: This artist I was working with, like most, had set prices per added character. However, I wanted some very tiny background characters who would be lower detail. Characters about the size of a pencil eraser. They wanted to charge full price for each character as though they were like those in detail in the foreground. We tried to negotiate a lower price, but I felt it was still ridiculously high for tiny background characters (around $25 each, if I recall). Because of this, I didn't return to this artist.

7. ... Want to be paid by the hour [X]

Look, I understand wanting to be paid minimum wage for your work. Art is a worthy job! Unfortunately, I have no way to monitor how many hours you actually worked on something. And what if you are slow on purpose? What if you personally are slow at art and not improving on your speed? I am not an employer at an animation studio or anything like that. This is too uncontrollable (and their estimated prices jump up to around $200+ USD for a scene that on average would otherwise cost me $50 USD). I think you should set your prices how you want then let people decide if that price suits them.

8. ... Want a signed contract [X]

Once again I understand not wanting to be scammed. I think the "pay upfront or pay after sketch" thing could protect you as an artist. Maybe you believe the contract comforts us both, but to me, it feels too suspicious or accusatory. No one wants to be treated as a scammer if they are sincere, right? It makes me too uncomfortable and the interaction too business-like. I prefer working with artists who are passionate about Transformers like me, which is why I only commission Transformers fan artists. That way we have real heart in the project. Let's not make it strictly about the money or a tight suspicious business interaction. If I see a contract to sign, I don't commission.

9. ... State "no posting or using the art" [X]

I commission art so that it can appear in my story game. If I pay an artist to make something that I can't show to others, it's a no go for me. I would like to reblog our commission or post it on social media while crediting the artist properly. Your clients are probably very excited to show their commissions (but the clients should never post the art without the credits and never sell that art for their own gain). I let my artists know what the art will be used for, but if I see a statement like this in a comm sheet, I don't reach out initially. It makes me think we'll have a conflict and aren't compatible.

10. ... Argue about how a scene or character should look [X]

This is an odd one, but here are two examples:

Client story: There were two characters in a scene. I asked if one of them could be smaller. The artist insisted they wanted the character to be larger (by this, I mean in height) because they liked big robots. I explained that this character wasn't particularly tall. For some reason, I had to push to get the character the height I wanted him for my story. It was unpleasant.

Client story: I like TFP Starscream without fangs. Some people like to add fangs. No biggie, until I'm the client paying you to draw him for me. They added fangs, I requested no fangs, it turned into an argument on why he should have fangs because he had jagged teeth in RID2015. Whatever it is, respect the client's preferences. Isn't the picture for them?

Please make your client happy and fulfill their requests.

11. ... Are rude or snappy! [X]

Client story: One time I reached out to an artist intending to do a huge commission. But I like to make the commission as fun as possible for my artists so I ask who their favourite Transformers characters are and if they prefer action or calm scenes. That way I can select a good scene for them off my giant list. And for this artist, I explained I had lots of options because it was a big story game. The artist demanded I just give my request, they said they just wanted to make money, then went on a rant that I was poorly organized. Well, we didn't do a commission and they missed out my $100 USD request.

Other suggestions to help you attract and retain clients:

Have someone proofread your comm sheet for typos and grammar (I have never turned someone down for this, it's just an idea)

Provide invoices that don't require addresses

Have a fun and positive comm sheet without passive aggressive statements. Do's and Don'ts are ok, but please don't seem bitter/fierce about something that may have happened in the past. Some terms and conditions really seemed to treat me like a troublemaker or threat!

Have a positive/non aggressive blog that avoids foul language

Appear open to discussion, flexible, and reasonable

These are just my suggestions: the suggestions of a client. It's up to you to accept or reject them as a paid artist. We all have our reasons. They are however, my secret reasons for not returning to certain artists or initially reaching out to them!

[canmom]

how much power does tech really use, compared to other shit?

my dash has been full of arguing about AI power consumption recently. so I decided to investigate a bit.

it's true, as the Ars Technica article argues, that AI is still only one fairly small part of the overall tech sector power consumption, potentially comparable to things like PC gaming. what's notable is how quickly it's grown in just a few years, and this is likely to be a limit to how much more it can scale.

I think it is reasonable to say that adding generative AI at large scale to systems that did not previously have generative AI (phones, Windows operating system etc.) will increase the energy cost. it's hard to estimate by how much. however, the bulk of AI energy use is in training, not querying. in some cases 'AI' might lead to less energy use, e.g. using an AI denoiser will reduce the energy needed to render an animated film.

the real problem being exposed is that most of us don't really have any intuition for how much energy is used for what. you can draw comparisons all sorts of ways. compare it to the total energy consumption of humanity and it may sound fairly niche; compare it to the energy used by a small country (I've seen Ireland as one example, which used about 170TWh in 2022) and it can sound huge.

but if we want to reduce the overall energy demand of our species (to slow our CO2 emissions in the short term, and accomodate the limitations of renewables in a hypothetical future), we should look at the full stack. how does AI, crypto and tech compare to other uses of energy?

here's how physicist David McKay broke down energy use per person in the UK way back in 2008 in Sustainable Energy Without The Hot Air, and his estimate of a viable renewable mix for the UK.

('Stuff' represents the embedded energy of manufactured goods not covered by the other boxes. 'Gadgets' represents the energy used by electronic devices including passive consumption by devices left on standby, and datacentres supporting them - I believe the embodied energy cost of building them falls under 'stuff' instead.)

today those numbers would probably look different - populations change, tech evolves, etc. etc., and this notably predates the massive rise in network infrastructure and computing tech that the Ars article describes. I'm sure someone's come up with a more up-to-date SEWTHA-style estimate of how energy consumption breaks down since then, but I don't have it to hand.

that said, the relative sizes of the blocks won't have changed that much. we still eat, heat our homes and fly about as much as ever; electric cars have become more popular but the fleet is still mostly petrol-powered. nothing has fundamentally changed in terms of the efficiency of most of this stuff. depending where you live, things might look a bit different - less energy on heating/cooling or more on cars for example.

how big a block would AI and crypto make on a chart like this?

per the IEA, crypto used 100-150TWh of electricity worldwide in 2022. in McKay's preferred unit of kWh/day/person, that would come to a worldwide average of just 0.04kWh/day/person. that is of course imagining that all eight billion of us use crypto, which is not true. if you looked at the total crypto-owning population, estimated to be 560 million in 2024, that comes to about 0.6kWh/day/crypto-owning person for cryptocurrency mining [2022/2024 data]. I'm sure that applies to a lot of people who just used crypto once to buy drugs or something, so the footprint of 'heavier' crypto users would be higher.

I'm actually a little surpised by this - I thought crypto was way worse. it's still orders of magnitude more demanding than other transaction systems but I'm rather relieved to see we haven't spent that much energy on the red queen race of cryptomining.

the projected energy use of AI is a bit more vague - depending on your estimate it could be higher or lower - but it would be a similar order of magnitude (around 100TWh).

SEWTHA calculated that in 2007, data centres in the USA added up to 0.4kWh/day/person. the ars article shows worldwide total data centre energy use increasing by a factor of about 7 since then; the world population has increased from just under 7 billion to nearly 8 billion. so the amount per person is probably about a sixfold increase to around 2.4kWh/day/person for data centres in the USA [extrapolated estimate based on 2007 data] - for Americans, anyway.

however, this is complicated because the proportion of people using network infrastructure worldwide has probably grown a lot since 2007, so a lot of that data centre expansion might be taking place outside the States.

as an alternative calculation, the IEA reports that in 2022, data centres accounted for 240-340 TWh, and transmitting data across the network, 260-360 TWh; in total 500-700TWh. averaged across the whole world, that comes to just 0.2 kWh/day/person for data centres and network infrastructure worldwide [2022 data] - though it probably breaks down very unequally across countries, which might account for the huge discrepancy in our estimates here! e.g. if you live in a country with fast, reliable internet where you can easily stream 4k video, you will probably account for much higher internet traffic than someone in a country where most people connect to the internet using phones over data.

overall, however we calculate it, it's still pretty small compared to the rest of the stack. AI is growing fast but worldwide energy use is around 180,000 TWh. humans use a lot of fucking energy. of course, reducing this is a multi-front battle, so we can still definitely stand to gain in tech. it's just not the main front here.

instead, the four biggest blocks by far are transportation, heating/cooling and manufacturing. if we want to make a real dent we'd need to collectively travel by car and plane a lot less, insulate our houses better, and reduce the turnover of material objects.

[captainxtra]

Major June 26th Political Update!

So, Paxton vs Free Speech Coalition and Birthright Citizenship + National Injunction cases will be settled most certainly tomorrow.

Keep an eye and ear out.

------

Now onto the Spending Bill news!

Here is a combined summary of all the updates from the Politico articles I checked:

The Republican-led "megabill" (also referred to as the "One Big Beautiful Bill") is facing numerous challenges and undergoing significant revisions due to rulings by Senate Parliamentarian Elizabeth MacDonough and internal disagreements within the GOP. Republicans are pushing to pass this comprehensive domestic policy bill—which includes tax, defense, energy, immigration, and health care provisions—before a self-imposed July 4 deadline, using the budget reconciliation process to bypass a Democratic filibuster.

Key Challenges and Parliamentarian Rulings:

Medicaid Provisions: The parliamentarian has dealt a major blow to the GOP's health care plans by ruling that several key Medicaid provisions cannot pass with a simple majority. This includes a plan to reduce Medicaid costs by cracking down on state provider taxes, which was expected to generate substantial savings, and proposals to exclude undocumented residents from Medicaid. These rulings have created an estimated $250 billion budgetary shortfall for Republicans.

Civil Service and Pension Reform: Initial proposals to restructure the federal workforce, such as giving federal employees an "at-will" designation and billing unions for official activities, were ruled out of order by the parliamentarian. Republicans are now exploring changes to their pension reform plan, proposing to increase federal employees' retirement contributions to 15.6 percent (from 9.4 percent) to help fund the megabill, though members of Congress, their staff, and federal law enforcement would be exempt. Democrats and unions strongly oppose these changes.

CFPB Funding: Senate Banking Republicans are proposing to cap the Consumer Financial Protection Bureau's (CFPB) funding at 6.5 percent of the Federal Reserve's operating budget, a reduction from the current 12 percent. An initial attempt to eliminate CFPB funding entirely was deemed ineligible. This new proposal awaits parliamentarian approval.

AI Moratorium: The parliamentarian has requested Senator Ted Cruz rewrite a proposed 10-year moratorium on enforcing state artificial intelligence (AI) laws, clarifying its scope of funding. While the measure is tied to a new $500 million fund for AI infrastructure, concerns remain that it could implicitly affect the entire $42 billion broadband program. The AI moratorium has also created divisions within the Republican party.

"Revenge Tax": Treasury Secretary Scott Bessent has asked lawmakers to remove a projected $52 billion "revenge tax" from the megabill, stating it is no longer necessary due to a "new understanding" with other developed countries regarding a global tax agreement, where central taxes will not apply to U.S. companies.

Overruling Parliamentarian: Despite calls from some conservative Republicans to do so, Senate Majority Leader John Thune has stated that the Senate will not move to overrule Parliamentarian Elizabeth MacDonough, emphasizing the importance of preserving Senate norms.

Approved Provisions:

SNAP Plan: The parliamentarian has approved a modified version of the Republican plan to shift some costs of the Supplemental Nutrition Assistance Program (SNAP) onto states. This is a significant win for Republicans, as it allows them to maintain a crucial $41 billion spending cut for the bill.

Internal Republican Divisions & Negotiations:

Medicaid Moderates: The proposed reduction of provider taxes in Medicaid expansion states is facing a "Medicaid-fueled mutiny" from moderate senators like Susan Collins, who find the proposed $15 billion rural hospital stabilization fund "inadequate." Some are threatening to block floor debate without more clarity on Medicaid changes.

SALT Deduction: Blue-state Republicans have rejected a Treasury offer to raise the cap on the State and Local Tax (SALT) deduction to $40,000, calling it "unrealistic." This remains a key unresolved issue.

Public Land Sales: Five House Republicans have declared a "red line" against including public land sales in the megabill, threatening to vote against it. They view it as a "grave mistake" and a "poison pill." This provision also faces strong Senate opposition.

Leadership Meetings & Deadlines:

Senate Majority Leader John Thune met with President Donald Trump at the White House as Republicans work against their self-imposed July 4 deadline. There is skepticism among some in the Senate GOP about whether they can get the bill to the President by this deadline. An initial vote to begin debate on the bill is now not expected before Saturday.

Kansas Governor Laura Kelly (D) anticipates that Senator Jerry Moran (R-Kan.), despite his objections to certain bill provisions, will ultimately vote with his party due to pressure from the comprehensive nature of the megabill.

The combination of parliamentarian rulings, internal Republican dissent, and complex negotiations across various policy areas is making the passage of the "megabill" a challenging and fluid process.

------

It seems that the bill is still being worked on so contact your Senators on troublesome provisions, reminding them to invoke the Byrd Rule against troublesome provisions should a procedural vote come before a new text is released.

General Page to call on the bill:

Page to call against Section 203, the provision that targets courts by requiring an expensive bond be payed to sue the government:

Page on AI regulation Ban:

------

Oh and congrats to Zohran Mamdani!

[solar-winds-ttrpg]

Looking for illustrators for my space fantasy TTRPG

Hello! I am looking for artists, especially local artists in or from the Nordics, to work with in creating images and illustrations for my Tabletop Role-Playing Game, Solar Winds. Any artwork provided will be for commercial use, fully credited and payed appropriately. Rates, stylisation wishes, and further details are listed below. If you are not interested, but know someone who might be, please share this with them!

Solar Winds is a space fantasy TTRPG that draws inspiration from various Nordic myths and folklore, with an especial focus on Finnish and Sámi folklore, as well as other stories such as Star Wars. It focuses on roleplay-driven storytelling while providing an elegant structure to combat, and allowing players to traverse both stars and a multitude of climates and environments. Think heroic or epic fantasy, with the addition of being set in space and containing starships.

Artists chosen for the project will be paid 50% up front, 50% after completion. I have a total budget of about 15 000kr (NOK) (~1 300€). The project is funded by TVIBITstigen (https://tvibit.net/stigen). The budget is split between the artpieces like this:

2x Landscapes / full scenes, 3000kr each

6x Characters, 1000kr each

2x Vehicle designs, 1000kr each

Smaller sketches or concepts for weapons, armour, etc., about 500kr each (These are my estimations for how much I'm going to pay for the different kinds of artwork, though final price will be agreed upon with each artist.)

If you are interested, you should take contact! To do so, you send your portfolio, or a link to your portfolio, to my e-mail:TheRightDragonEye@gmail.com. (You should make sure your link does not require users to have an account to see your work.) Please also mention what kinds of artpieces you would be interested in / best suited to create, as this is big project with a variety of scenes.

I'm most interested in semi-realistic artstyles, semi-stylised artstyles, and paintings (digital or otherwise). While the artworks would need to be delivered digitally, I accept both digital and traditional artstyles for this project. You will also have an advantage if your portfolio includes some of the following subjects:

Process images or timelapses (to root out AI)

Science fiction

Diverse landscapes

Various insectoid humanoids (bug people) (this one is a massive plus)

Mechanical humanoids (droids, exos, robots, etc.)

Vehicles (starships, hoverbikes, walkers, cars, etc.) (especially if self-designed)

City streets, megacities

Action, combat, war

Post-apocalypse

Armour and weapon designs (especially if self-designed)

A mixture of fantasy and technology

Cryptids, folkloric entities

Northern lights

In addition, while anyone can apply here, I am especially looking for artists in or from the Nordic region, especially Sámi or Finnish artists, as this project takes a lot of inspiration from Sámi and Finnish folklore in particular. If you have another indigenous background, that is also an advantage!

I will accept NO AI-generated or assisted "art".

If you have any questions or comments, please ask away <3

[kynndr3dd-synn3]

0BAMA ‘CRIME FAMILY’ : What's The TRUE ORIGIN Of Their Wealth⁉️

Barack and Michelle Obama’s net worth is estimated between $70M and $135M - a sharp rise from just $1.5M in 2008. How did they amass such wealth?

🔴 Between 2005 and 2016, the couple earned $20.5M in salaries and book royalties, per Forbes.

🔴 By his 2009 inauguration, Obama held $1.1M-$5.1M in US Treasury bills.

🔴 In 2017, they secured a $65M joint book deal.

🔴 Barack made $1.2M from just three Wall Street speaking engagements, with estimates of $400K per event - equal to his presidential salary.

🔴 Michelle reportedly charges $225K per speech, with 2018 book tour tickets ranging from $307 to $4,070 (Business Insider).

🔴 In 2018, they founded Higher Ground Productions and signed a $50M Netflix deal.

🔴 In 2019, they inked an exclusive podcast deal with Spotify, followed by a first-look deal with Audible in 2022.

POLITICAL DYNASTY GRIFTERS...

While mainstream outlets insist there’s nothing unusual about their fortune, a Kogod School of Business projection suggests the couple could earn $242.5M by the 2030s.

However, US conservatives on X are urging Elon Musk's DOGE community to INVESTIGATE the true extent of the Obamas' wealth, citing their lavish lifestyle and luxury properties.

Wall Street analyst and charity fraud expert Charles Ortel suspects the Obamas’ philanthropic ventures mimic the Clinton Foundation’s alleged fraud:

“The IRS… has never disciplined the Obamas for corporate mining naked partisan and personal enrichment schemes inside a ‘public charity,’ which is strictly and clearly illegal,” he believes.

Ortel notes the Obama-era FBI & DOJ failed to complete investigations into the Clinton Foundation’s alleged pay-to-play schemes. He believes the Clintons and Obamas pushed Russiagate and post-2020 lawfare against Trump to prevent him from using presidential powers to thoroughly investigate them.

The image is generated by AI

👍 @geopolitics_live

[inbabylontheywept]

So, just some Fermi numbers for AI: I'm going to invent a unit right now, 1 Global Flop-Second. It's the total amount of computation available in the world, for one second. That's 10^21 flops if you're actually kind of curious. GPT-3 required about 100 Global Flop-Seconds, or nearly 2 minutes. GPT-4 required around 10,000 Global Flop-Seconds, or about 3 hours, and at the time, consumed something like 1/2000th the worlds total computational capacity for a couple of years. If we assume that every iteration just goes up by something like 100x as many flop seconds, GPT-5 is going to take 1,000,000 Global Flop-Seconds, or 12 days of capacity. They've been working on it for a year and a half, which implies that they've been using something like 1% of the world's total computational capacity in that time.

So just drawing straights lines in the guesses (this is a Fermi estimation), GPT-6 would need 20x as much computing fraction as GPT-5, which needed 20x as much as GPT-4, so it would take something like a quarter of all the world's computational capacity to make if they tried for a year and a half. If they cut themselves some slack and went for five years, they'd still need 5-6%.

And GPT-7 would need 20x as much as that.

OpenAI's CEO has said that their optimistic estimates for getting to GPT-7 would require seven-trillion dollars of investment. That's about as much as Microsoft, Apple, and Google combined. So, for limiting factors involved are... GPT-6: Limited by money. GPT-6 doesn't happen unless GPT-5 can make an absolute shitload. Decreasing gains kill this project, and all the ones after that. We don't actually know how far deep learning can be pushed before it stops working, but it probably doesnt' scale forever. GPT-7: Limited by money, and by total supply of hardware. Would need to make a massive return on six, and find a way to actually improve hardware output for the world. GPT-8: Limited by money, and by hardware, and by global energy supplies. Would require breakthroughs in at least two of those three. A world where GPT-8 can be designed is almost impossible to imagine. A world where GPT-8 exists is like summoning an elder god. GPT-9, just for giggles, is like a Kardeshev 1 level project. Maybe level 2.