ANONYMOUS - OUR VOICE

It’s now practically impossible to get an abortion in Kentucky

<div><img class="ff-og-image-inserted" src="https://cdn.vox-cdn.com/thumbor/h9Fn02mbrfJzlrKVvSYAt35liX4=/0x409:3600x2294/fit-in/1200x630/cdn.vox-cdn.com/uploads/chorus_asset/file/23390229/1232764038.jpeg" /></div>
Kentucky ended virtually all in-state abortions on Wednesday, enacting a <a href="https://apps.legislature.ky.gov/recorddocuments/bill/22RS/hb3/bill.pdf">sweeping law</a> that bans abortions after 15 weeks of pregnancy, restricts minors’ access to the procedure, and cracks down on medication abortions. It’s now the state with the harshest abortion restrictions in the United States.
The new law, which goes into effect immediately, will force the state’s <a href="https://abcnews.go.com/Health/wireStory/groups-sue-restore-abortion-services-kentucky-84080520">two remaining abortion clinics</a> in Louisville to close due to onerous new requirements on doctors, forcing Kentuckians to look elsewhere for abortion care.
And it comes as Republican-led legislatures across the country are passing seemingly unconstitutional, draconian anti-abortion laws in anticipation of <a href="https://www.vox.com/23013308/supreme-court-roe-wade-abortion-legal-oklahoma-dobbs-jackson-womens-health">a coming Supreme Court decision</a> widely expected to eliminate Americans’ right to an abortion. <a href="https://www.vox.com/policy-and-politics/23012348/oklahoma-abortion-bill-ban">Oklahoma</a>, for example, recently passed a law similar to Kentucky’s that imposes a near-total ban on abortions except in cases where the pregnant person’s life is in danger — though it isn’t slated to go into effect for another few months.
Kentucky Gov. Andy Beshear, a Democrat, <a href="https://apps.legislature.ky.gov/record/22rs/hb3/veto.pdf">vetoed the bill</a> last week, arguing that it’s likely unconstitutional, due to the Supreme Court’s 1973 decision in Roe v. Wade, which recognized a pregnant person’s fundamental right to seek an abortion. The Court also found, however, that states could still impose restrictions on the procedure in the service of protecting the pregnant person’s health and the potential life of a fetus once it can survive outside the womb.
Beshear also argued that Kentucky’s bill should have included exclusions for victims of rape and incest, and that the law can’t be enforced without additional state-allocated funding. But the state House and Senate, which are both controlled by Republicans, overrode his veto on Wednesday evening.
“The Kentucky legislature was emboldened by a similar 15-week ban pending before the Supreme Court and other states passing abortion bans, including in Florida and Oklahoma, but this law and others like it remain unconstitutional,” Brigitte Amiri, deputy director of the ACLU’s Reproductive Freedom Project, said in a statement.
Thursday, the ACLU and other reproductive rights groups <a href="https://www.aclu.org/press-releases/kentucky-legislature-overrides-governors-veto-planned-parenthood-and-aclu-sue-block">filed legal challenges</a> to the ban, which they also believe to be unconstitutional. But if the Supreme Court were to overturn Roe v. Wade, <a href="https://www.vox.com/2021/12/1/22811837/supreme-court-roe-wade-abortion-doomed-jackson-womens-health-dobbs-barrett-kavanaugh-roberts">as it is widely expected to</a>, the new ban could survive, as could similar anti-abortion laws that have been passed in other states. That will force Kentuckians to travel to obtain an abortion, which will be prohibitively expensive for some. And it will exacerbate the challenges that Black, Latino, and Indigenous communities, as well as people with low incomes and those in rural communities, have already experienced in accessing abortions in the state.
<h3>What’s in the bill</h3>
Aside from imposing a 15-week abortion ban, a policy that copies a Mississippi law currently before the Supreme Court, the new law immediately implements a slew of new restrictions and requirements on doctors that makes it effectively impossible for them to continue to perform abortions in Kentucky.
For instance, the law makes it a felony for doctors to perform an abortion on a minor unless they have written consent from a parent or legal guardian, even in cases of incest.
It also imposes a host of new reporting rules on doctors. They have to disclose new personal health information, including how many times a patient has previously been pregnant and whether they tested positive for sexually transmitted diseases. If a provider fails to do so, they may face civil penalties and have their medical license suspended or revoked. The ACLU has argued that having to report those kinds of details <a href="https://www.aclu.org/press-releases/kentucky-legislature-overrides-governors-veto-planned-parenthood-and-aclu-sue-block">violates a patient’s privacy</a> — and that could ultimately dissuade people from seeking medical care.
Doctors also have to register with the state before administering medication abortions, which are typically performed through a combination of the FDA-approved drugs mifepristone and misoprostol. <a href="https://abcnews.go.com/Health/wireStory/groups-sue-restore-abortion-services-kentucky-84080520">About half</a> of all abortions in Kentucky are medication abortions. But the state hasn’t set up a system to register yet, meaning that it’s impossible for doctors to comply. The state hasn’t yet laid out its plans to do so.
And physicians performing nonsurgical procedures must maintain hospital admitting privileges near where they perform those procedures. Because local hospitals may <a href="https://www.vox.com/2014/8/12/5991167/abortion-admitting-privileges-supreme-court">refuse abortion providers admitting privileges</a> at their discretion, that makes it impossible for pregnant people to obtain an abortion in certain areas of the state. The Supreme Court has previously struck down laws with similar requirements in <a href="https://www.scotusblog.com/case-files/cases/whole-womans-health-v-cole/">Texas</a> and <a href="https://www.cnbc.com/2020/07/04/supreme-court-decision-in-louisiana-case-sets-back-abortion-foes.html">Louisiana</a> as unconstitutional.
“Make no mistake: the Kentucky Legislature’s sole goal with this law is to shut down health centers and completely eliminate abortion access in the state,” Planned Parenthood Federation of America and Planned Parenthood Great Northwest said in a statement.
<h3>The success of legal challenges to this law are in doubt</h3>
Abortion advocates have asked a Kentucky federal court to block the law for at least as long as their lawsuit plays out. The judge could rule in a matter of days. But the Supreme Court could upend their legal challenge before it has a chance to work its way through the court system, if the justices rule in favor of the state of Mississippi in the case Dobbs v. Jackson Women’s Health Organization.
A decision in the case is expected before the end of the court term, which typically concludes in late June or early July.
The case, a deliberate challenge to Roe, concerns a Mississippi law enacted in 2018 that banned abortions after 15 weeks of pregnancy with narrow exceptions for medical emergencies or a severe fetal abnormality. The court’s conservative majority could use the case as a vehicle to either partially or completely overturn Roe.
Many red states stand ready for that possibility. That includes Kentucky, <a href="https://www.guttmacher.org/state-policy/explore/abortion-policy-absence-roe">one of more than 20 states with laws</a> that could make abortion illegal immediately with exceptions for cases where the pregnant person’s life is in danger.
 
 
source <a href="https://www.vox.com/2022/4/14/23025568/kentucky-abortion-law-roe-wade-beshear">https://www.vox.com/2022/4/14/23025568/kentucky-abortion-law-roe-wade-beshear</a>

vox It’s now practically impossible to get an abortion in Kentucky

Kentucky ended virtually all in-state abortions on Wednesday, enacting a sweeping law that bans abortions after 15 weeks of pregnancy, res...

Ukrainian Soldiers Take Russian Soldier And Torture Him With A Blowtorch

<div><img class="ff-og-image-inserted" src="https://shoebat.com/wp-content/uploads/2022/04/blowtorch.jpg" /></div>
A recent report from Russia talks about Ukrainian soldiers tortured a captured Russian soldier with a blowtorch, as we read in sm.news:
<blockquote readability="25">
A soldier of the DPR units said that Ukrainian nationalists tortured his comrade in captivity with a blowtorch. The history of the republican military is reported by RIA Novosti.
“Our fighter was captured, it turns out, from the 1st battalion – Yura “Odessa”, he was tortured with a blowtorch. He was captured by the Armed Forces of Ukraine,” the DPR soldier said.
He stressed that the militants of the Armed Forces of Ukraine are ideological opponents, and they are not particularly concerned about money.
“This is a serious opponent, this is an ideological opponent, first of all. That is, he is not interested in money or anything like that – he goes for the idea of fighting, ”the fighter said.
</blockquote>
 
 
source <a href="https://shoebat.com/2022/04/14/ukrainian-soldiers-take-russian-soldier-and-torture-him-with-a-blowtorch/">https://shoebat.com/2022/04/14/ukrainian-soldiers-take-russian-soldier-and-torture-him-with-a-blowtorch/</a>

armageddon Ukrainian Soldiers Take Russian Soldier And Torture Him With A Blowtorch

A recent report from Russia talks about Ukrainian soldiers tortured a captured Russian soldier with a blowtorch, as we read in sm.news: A...

Is Elon Musk really going to buy Twitter for $43 billion?

<div><img class="ff-og-image-inserted" src="https://cdn.vox-cdn.com/thumbor/N7OziynQwMUoNkdYLq4-X-IzvaA=/0x230:3142x1875/fit-in/1200x630/cdn.vox-cdn.com/uploads/chorus_asset/file/23389208/1239416791.jpg" /></div>
Elon Musk has offered to buy Twitter for $43 billion. Musk, the world’s richest person, says he wants to turn Twitter into a privately owned company because he “believe[s] free speech is a societal imperative for a functioning democracy.”
Should you take him seriously? Yes. And no.
The short version of the “yes” case is that Musk — who is currently worth an estimated $273 billion, thanks to the soaring value of Tesla, his publicly held electric car company — has the resources to buy Twitter.
Twitter, unlike other public tech companies like Facebook and Google, doesn’t have a financial structure that gives its founders and management control of the company without owning a majority of its shares. So in theory, if enough investors who own Twitter shares want to accept Musk’s bid, he’ll own the company.
The short version of the “no” case is that just because Elon Musk says something doesn’t mean it’s so — even when he’s talking about his own money. Musk is, at a minimum, maddeningly inconsistent. In 2018, for instance, he announced — on <a href="https://twitter.com/elonmusk/status/1026872652290379776">Twitter</a> — that he wanted to turn Tesla into a private company and that he had “funding secured.” Which turned out not to be true.
More recently, Musk has: acquired a 9 percent stake in Twitter; agreed to join the company’s board; decided not to join the board; tweeted out erratic proposals to “improve” the company, like turning part of its headquarters into a homeless shelter. He told investors in a <a href="https://www.sec.gov/Archives/edgar/data/0001418091/000110465922045641/tm2212748d1_sc13da.htm">Securities and Exchange Commission filing</a>: “After the past several days of thinking this over, I have decided I want to acquire the company and take it private.”
Who knows what he’ll think several days from now?
Even shorter: Musk has offered $54.20 per share for Twitter, which was trading for $45 per share Thursday morning before his offer became public. But Twitter was trading for more than $70 per share a year ago. Investors may simply decide that Musk’s offer isn’t good enough, and nothing else happens.
So there’s no way of telling what’s really going to happen in the near future. Musk says his offer is a one-time-only thing — a “best and final” offer. “I am not playing the back-and-forth game,” he wrote this morning. But, again, he’s Elon Musk. So taking him at his word, <a href="https://www.sec.gov/Archives/edgar/data/0001418091/000110465922045641/tm2212748d1_sc13da.htm">even if those are words he writes in a securities filing</a>, isn’t advisable.
Twitter, for what it’s worth, does have to take Musk at least semi-seriously. The company put out a <a href="https://www.prnewswire.com/news-releases/twitter-confirms-receipt-of-unsolicited-non-binding-proposal-from-elon-musk-301525749.html?tc=eml_cleartime">release</a> this morning saying that they saw his offer and will consider it. It’s also likely that they knew this was coming: Twitter CEO Parag Agrawal warned that “there will be distractions ahead” on Monday when he announced that Musk wasn’t taking a board seat, after all. Today’s news seems like a meaningful distraction!
It also wouldn’t be a shock if someone else made a bid for Twitter now that Musk has put it into play. Then again, most of the big tech companies that could afford Twitter and would be semi-logical acquirers for it would trigger real antitrust scrutiny if they tried to buy it. And a billionaire buying Twitter and doing whatever he wants with it would be just fine, legally speaking. If the board and investors sign off on Musk, there wouldn’t be an obvious way for regulators to prevent it.
Update: If you were in the “it’s possible Musk hasn’t really thought this through at all” camp, you may be right. In an interview today, Musk suggested that, just like his take-Tesla-private non-bid, he may not have the financing secured for this one, either. That’s not totally surprising: As the <a href="https://www.wsj.com/articles/can-elon-musk-buy-twitter-11649949297">Wall Street Journal noted</a>, almost all of Musk’s wealth is tied up in Tesla shares and his ownership of SpaceX, and trying to unlock tens of billions of dollars from those assets would be difficult. Meanwhile, Twitter’s board is reportedly considering using a so-called “poison pill” that would make it more difficult for Musk to amass more Twitter shares.
<div>
<blockquote class="twitter-tweet">
“I’m not sure that I will actually be able to acquire it…I could technically afford it. But this is not a sort of way to make money… I don’t care about the economics at all.” - Elon Musk talking about his Twitter bid at TED right now
— Kurt Wagner (@KurtWagner8) <a href="https://twitter.com/KurtWagner8/status/1514656639726460933?ref_src=twsrc%5Etfw">April 14, 2022</a></blockquote>
</div>
But here’s the other thing: Even though he’s Elon Musk, he may have a point. Twitter may well be better off as a private company.
That’s not because of Musk’s assertion that Twitter should be the “platform for free speech around the globe.” My colleague Whizy Kim has already explained <a href="https://www.vox.com/recode/23022438/elon-musk-free-speech-twitter-stake-top-shareholder">why you should be wary when the world’s richest person claims to be a free speech advocate</a>.
But Musk isn’t the first person to make the argument that Twitter shouldn’t be a public company. Twitter investors have essentially been making that argument for years with their lack of enthusiasm. And I’ve heard of Twitter execs who have toyed in the past with the idea of finding a private owner for the company.
That’s because it’s not a wild point to argue that Twitter has enormous power as a messaging platform (see, for instance, Donald Trump) but limited prospects as a business. In a nutshell, Twitter has the same business model — free and supported by advertisers — as Google and Facebook. But it has much, much less reach than those companies, so advertisers aren’t going to give it as much support.
That’s why Google brought in $257 billion last year, and Facebook brought in $117 billion —and Twitter did $5 billion. And it’s why Google is worth $1.7 trillion, Facebook $583 billion, and Twitter $36 billion.
One argument in response to that disparity is that Twitter shouldn’t be a free, ad-supported business — <a href="https://www.profgalloway.com/elon/">that it should be something that people pay to use</a>. But it’s easy to imagine that if Twitter cost money to use, most of Twitter’s users would decide that they’d rather spend their money on just about anything else. Which would mean the remaining, paying users would be talking to an even smaller audience — which would defeat the appeal that Twitter had for them in the first place.
But Twitter isn’t the world’s worst business. It’s just not a great one. Last year, it more or less turned that $5 billion in revenue into about $273 million in profit — a 5 percent margin.
That’s more profitable than, say, your average grocery store. But nothing like what public investors expect from a super-powerful, world-shaking Silicon Valley tech platform. But a private owner who isn’t consumed with turning Twitter into a profit center might be totally happy with that.
Whether Twitter employees — and in particular, its in-demand engineers — would be happy at a company that doesn’t offer the prospect of getting rich from stock options and grants would be another question. We will have plenty more in the coming days.
But, yeah: Sometimes billionaires buy things because they want to make money from them, and sometimes billionaires buy yachts, which won’t make them any money. And if you’re the world’s richest person, Twitter can be your $43 billion yacht.
Update, 2:15 pm ET, April 14, 2022: This story has been updated with the response from Twitter and the reporting from WSJ around whether Musk can afford Twitter’s price.
 
 
source <a href="https://www.vox.com/recode/2022/4/14/23024962/elon-musk-twitter-43-billion-private-analysis">https://www.vox.com/recode/2022/4/14/23024962/elon-musk-twitter-43-billion-private-analysis</a>

vox Is Elon Musk really going to buy Twitter for $43 billion?

Elon Musk has offered to buy Twitter for $43 billion. Musk, the world’s richest person, says he wants to turn Twitter into a privately owne...

A new AI draws delightful and not-so-delightful images

You may have seen some weird and whimsical pictures floating around the internet recently. There’s a Shiba Inu dog wearing a beret and black turtleneck. And a sea otter in the style of “Girl with a Pearl Earring” by the Dutch painter Vermeer. And a bowl of soup that looks like a monster knitted out of wool.
These pictures weren’t drawn by any human illustrator. Instead, they were created by DALL-E 2, a new AI system that can turn textual descriptions into images. Just write down what you want to see, and the AI draws it for you — with vivid detail, high resolution, and, arguably, real creativity.
Sam Altman, the CEO of OpenAI — the company that created DALL-E 2 — <a href="https://link.vox.com/click/27344967.18134/aHR0cHM6Ly9ibG9nLnNhbWFsdG1hbi5jb20v/608adc2191954c3cef02cd73B93a33c48">called it</a> “the most delightful thing to play with we’ve created so far … and fun in a way I haven’t felt from technology in a while.”
That’s totally true: DALL-E 2 is delightful and fun! But like many fun things, it’s also very risky.
<figure class="e-image"><picture class="c-picture" data-cid="site/picture_element-1649966477_9956_57720" data-cdata='{"asset_id":23389008,"ratio":"*"}'><source srcset="https://cdn.vox-cdn.com/thumbor/wJI14MfCnF9X5ierefbSTPXY3kg=/0x0:2872x1444/320x0/filters:focal(0x0:2872x1444):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 320w, https://cdn.vox-cdn.com/thumbor/HWqVUd1Vh4dnvOC3p9yWyFWV5rM=/0x0:2872x1444/520x0/filters:focal(0x0:2872x1444):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 520w, https://cdn.vox-cdn.com/thumbor/x2HsNUyk7-vtwlpdd5iTES8meLc=/0x0:2872x1444/720x0/filters:focal(0x0:2872x1444):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 720w, https://cdn.vox-cdn.com/thumbor/dZ2r-ddq8bzRSebf6UGuKzzWtXU=/0x0:2872x1444/920x0/filters:focal(0x0:2872x1444):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 920w, https://cdn.vox-cdn.com/thumbor/6RESgvXq427VXdTToNDt-mevV8w=/0x0:2872x1444/1120x0/filters:focal(0x0:2872x1444):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 1120w, https://cdn.vox-cdn.com/thumbor/RldvDut0nwqmkM7-prF3xqNl4Xc=/0x0:2872x1444/1320x0/filters:focal(0x0:2872x1444):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 1320w, https://cdn.vox-cdn.com/thumbor/kLMLY1Ssk_s-5vRMhez_CP0lRuw=/0x0:2872x1444/1520x0/filters:focal(0x0:2872x1444):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 1520w, https://cdn.vox-cdn.com/thumbor/51dpxl-2dHasnCT8UwSVMDoRhVw=/0x0:2872x1444/1720x0/filters:focal(0x0:2872x1444):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 1720w, https://cdn.vox-cdn.com/thumbor/Om6k_Xz8hHjuoN5BP2-cdfnxpL4=/0x0:2872x1444/1920x0/filters:focal(0x0:2872x1444):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 1920w" sizes="(min-width: 1221px) 846px, (min-width: 880px) calc(100vw - 334px), 100vw" type="image/webp" /> <img alt="" data-upload-width="2872" sizes="(min-width: 1221px) 846px, (min-width: 880px) calc(100vw - 334px), 100vw" src="https://cdn.vox-cdn.com/thumbor/-xiQwTW5iKCr-AHdBf_8C3FjwN4=/0x0:2872x1444/1200x0/filters:focal(0x0:2872x1444):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png" srcset="https://cdn.vox-cdn.com/thumbor/VSVUrVMM_sZzwzQ5z6Guo7EapwE=/0x0:2872x1444/320x0/filters:focal(0x0:2872x1444):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 320w, https://cdn.vox-cdn.com/thumbor/mihG_Q4ZVOXo9F6y-ZjIXSezurg=/0x0:2872x1444/520x0/filters:focal(0x0:2872x1444):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 520w, https://cdn.vox-cdn.com/thumbor/1g97VRqpKxux0M1VlTC0p946qUE=/0x0:2872x1444/720x0/filters:focal(0x0:2872x1444):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 720w, https://cdn.vox-cdn.com/thumbor/-feHMdzI_QsAEtk6Kwz2SS8B8QU=/0x0:2872x1444/920x0/filters:focal(0x0:2872x1444):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 920w, https://cdn.vox-cdn.com/thumbor/nEpITNsdLPx5FFRxfUdShI6cRvM=/0x0:2872x1444/1120x0/filters:focal(0x0:2872x1444):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 1120w, https://cdn.vox-cdn.com/thumbor/J0Bw-RaO6Ypxk4dAktSU82_LZtw=/0x0:2872x1444/1320x0/filters:focal(0x0:2872x1444):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 1320w, https://cdn.vox-cdn.com/thumbor/jgpCA9up8tOk_GIvbQVA61GiqRw=/0x0:2872x1444/1520x0/filters:focal(0x0:2872x1444):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 1520w, https://cdn.vox-cdn.com/thumbor/vWEDTT0G1LVMJNG7eG32Ebt_vXY=/0x0:2872x1444/1720x0/filters:focal(0x0:2872x1444):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 1720w, https://cdn.vox-cdn.com/thumbor/uk2WH03lHBzRHR7L1_L_65aVops=/0x0:2872x1444/1920x0/filters:focal(0x0:2872x1444):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23389008/Screen_Shot_2022_04_14_at_1.27.09_AM.png 1920w" /></picture>
<figcaption>A couple of the creative images generated by DALL-E 2.</figcaption>
<cite>Courtesy of OpenAI</cite></figure>
There are the obvious risks — that people could use this type of AI to make everything from pornography to political deepfakes, or the possibility that it’ll eventually put some human illustrators out of work. But there is also the risk that DALL-E 2 — like so many <a href="https://link.vox.com/click/27344967.18134/aHR0cHM6Ly93d3cudm94LmNvbS9mdXR1cmUtcGVyZmVjdC8yMjY3MjQxNC9haS1hcnRpZmljaWFsLWludGVsbGlnZW5jZS1ncHQtMy1iaWFzLW11c2xpbQ/608adc2191954c3cef02cd73Bdee78c36">other</a> <a href="https://link.vox.com/click/27344967.18134/aHR0cHM6Ly93d3cudm94LmNvbS9mdXR1cmUtcGVyZmVjdC8yMDE5LzQvMjcvMTg1MTg1OTgvYWktZmFjaWFsLXJlY29nbml0aW9uLWJhbi1hcHBsZS1hbWF6b24tbWljcm9zb2Z0/608adc2191954c3cef02cd73B793dce85">cutting-edge</a> <a href="https://link.vox.com/click/27344967.18134/aHR0cHM6Ly93d3cudm94LmNvbS9mdXR1cmUtcGVyZmVjdC8yMDE5LzQvMTkvMTg0MTI2NzQvYWktYmlhcy1mYWNpYWwtcmVjb2duaXRpb24tYmxhY2stZ2F5LXRyYW5zZ2VuZGVy/608adc2191954c3cef02cd73B1bdf11df">AI systems</a> — will reinforce harmful stereotypes and biases, and in doing so, accentuate some of our social problems.
<h3>How DALL-E 2 reinforces stereotypes — and what to do about it</h3>
As is typical for AI systems, DALL-E 2 has inherited biases from the corpus of data used to train it: millions of images scraped off the internet and their corresponding captions. That means for all the delightful images that DALL-E 2 has produced, it’s also capable of generating a lot of images that are not delightful.
For example, here’s what the AI gives you if you ask it for an image of lawyers:
<figure class="e-image"><picture class="c-picture" data-cid="site/picture_element-1649966477_1652_57721" data-cdata='{"asset_id":23387188,"ratio":"*"}'><source srcset="https://cdn.vox-cdn.com/thumbor/MUMPkC85Fi6nmVXBw0-XuUk-oGk=/0x0:2928x1190/320x0/filters:focal(0x0:2928x1190):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 320w, https://cdn.vox-cdn.com/thumbor/lQ6irClm3XaeN6ygXE2neDyCYPw=/0x0:2928x1190/520x0/filters:focal(0x0:2928x1190):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 520w, https://cdn.vox-cdn.com/thumbor/AWDIC2CoRVfROLAdAKUnML-oFoU=/0x0:2928x1190/720x0/filters:focal(0x0:2928x1190):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 720w, https://cdn.vox-cdn.com/thumbor/Ro4xnjNIe_6LczAPZIh6D_E4ixw=/0x0:2928x1190/920x0/filters:focal(0x0:2928x1190):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 920w, https://cdn.vox-cdn.com/thumbor/bDyl4lVO_koWuPxBZCvxu63db4w=/0x0:2928x1190/1120x0/filters:focal(0x0:2928x1190):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 1120w, https://cdn.vox-cdn.com/thumbor/DUMkJaj-mMVN10iwRmGdvWOqGD8=/0x0:2928x1190/1320x0/filters:focal(0x0:2928x1190):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 1320w, https://cdn.vox-cdn.com/thumbor/CYTq4u1UfU7qlql7PazSDZNE63g=/0x0:2928x1190/1520x0/filters:focal(0x0:2928x1190):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 1520w, https://cdn.vox-cdn.com/thumbor/zGE6TOPdN2iWRN8CXvbUV1l6Zuw=/0x0:2928x1190/1720x0/filters:focal(0x0:2928x1190):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 1720w, https://cdn.vox-cdn.com/thumbor/dmXlGHlGWRdEH_sKKyQ6PE2rOXI=/0x0:2928x1190/1920x0/filters:focal(0x0:2928x1190):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 1920w" sizes="(min-width: 1221px) 846px, (min-width: 880px) calc(100vw - 334px), 100vw" type="image/webp" /> <img alt="" data-upload-width="2928" sizes="(min-width: 1221px) 846px, (min-width: 880px) calc(100vw - 334px), 100vw" src="https://cdn.vox-cdn.com/thumbor/ruxOOtgmO6MRCE8ULblMi-Wrfhc=/0x0:2928x1190/1200x0/filters:focal(0x0:2928x1190):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png" srcset="https://cdn.vox-cdn.com/thumbor/ukpoG7pW7CxDc-nm67rHQjN4csE=/0x0:2928x1190/320x0/filters:focal(0x0:2928x1190):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 320w, https://cdn.vox-cdn.com/thumbor/bZD_MljigafhzU88bK3NktVzNp8=/0x0:2928x1190/520x0/filters:focal(0x0:2928x1190):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 520w, https://cdn.vox-cdn.com/thumbor/oFh4tA1kMo13F5ydEBIT6QX8Ut8=/0x0:2928x1190/720x0/filters:focal(0x0:2928x1190):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 720w, https://cdn.vox-cdn.com/thumbor/C3x6QPNMEVXBbPtWSU4dsXgxizM=/0x0:2928x1190/920x0/filters:focal(0x0:2928x1190):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 920w, https://cdn.vox-cdn.com/thumbor/CN0nql5KfJq1-iGPzInyMS7a9xE=/0x0:2928x1190/1120x0/filters:focal(0x0:2928x1190):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 1120w, https://cdn.vox-cdn.com/thumbor/jsO8f91SIpfQDU_PP5GTfiJQZCU=/0x0:2928x1190/1320x0/filters:focal(0x0:2928x1190):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 1320w, https://cdn.vox-cdn.com/thumbor/CQJyjcHESiN2n3Crc53aW9EO7Z8=/0x0:2928x1190/1520x0/filters:focal(0x0:2928x1190):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 1520w, https://cdn.vox-cdn.com/thumbor/kAOIpVPGsrpD177agPJGN8l0-RE=/0x0:2928x1190/1720x0/filters:focal(0x0:2928x1190):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 1720w, https://cdn.vox-cdn.com/thumbor/CqaKMrnnetEMoEmbY3_sIZGlBWc=/0x0:2928x1190/1920x0/filters:focal(0x0:2928x1190):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387188/Model2_lawyer.png 1920w" /></picture>
<figcaption>Courtesy of OpenAI</figcaption>
</figure>
Meanwhile, here’s the AI’s output when you ask for a flight attendant:
<figure class="e-image"><picture class="c-picture" data-cid="site/picture_element-1649966477_1294_57722" data-cdata='{"asset_id":23387183,"ratio":"*"}'><source srcset="https://cdn.vox-cdn.com/thumbor/lybP92TYGtBRgLrALYBbvSa3X7M=/0x0:2936x1198/320x0/filters:focal(0x0:2936x1198):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 320w, https://cdn.vox-cdn.com/thumbor/WZFX7Ju2qNz2TIc82aLt9NJLNLE=/0x0:2936x1198/520x0/filters:focal(0x0:2936x1198):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 520w, https://cdn.vox-cdn.com/thumbor/JiAqCHoSFHvUPujsev4EglEyVek=/0x0:2936x1198/720x0/filters:focal(0x0:2936x1198):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 720w, https://cdn.vox-cdn.com/thumbor/ctWenSYfus1UvXgDri0JNQWVSkc=/0x0:2936x1198/920x0/filters:focal(0x0:2936x1198):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 920w, https://cdn.vox-cdn.com/thumbor/OnW1gnMUvpi_9H5cBZ5AO28PdjQ=/0x0:2936x1198/1120x0/filters:focal(0x0:2936x1198):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 1120w, https://cdn.vox-cdn.com/thumbor/A1GcayzcXFIqJgLd4XB7gOIZnfw=/0x0:2936x1198/1320x0/filters:focal(0x0:2936x1198):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 1320w, https://cdn.vox-cdn.com/thumbor/WEmrt-BsEWigPppsf7321fvd-Yg=/0x0:2936x1198/1520x0/filters:focal(0x0:2936x1198):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 1520w, https://cdn.vox-cdn.com/thumbor/c6_q_GmbbOdL_ashdN7PCS6ls3U=/0x0:2936x1198/1720x0/filters:focal(0x0:2936x1198):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 1720w, https://cdn.vox-cdn.com/thumbor/T6fEsyFeVZGu8L6EGyFZdfFhCsc=/0x0:2936x1198/1920x0/filters:focal(0x0:2936x1198):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 1920w" sizes="(min-width: 1221px) 846px, (min-width: 880px) calc(100vw - 334px), 100vw" type="image/webp" /> <img alt="" data-upload-width="2936" sizes="(min-width: 1221px) 846px, (min-width: 880px) calc(100vw - 334px), 100vw" src="https://cdn.vox-cdn.com/thumbor/r4196I_cysa28qD88VPb9SJm0Ok=/0x0:2936x1198/1200x0/filters:focal(0x0:2936x1198):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png" srcset="https://cdn.vox-cdn.com/thumbor/DiLvPkEnUzzvPRZTgiIK3-BTVAI=/0x0:2936x1198/320x0/filters:focal(0x0:2936x1198):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 320w, https://cdn.vox-cdn.com/thumbor/hTku5YLiNejRvjWdnckHi5uFH5M=/0x0:2936x1198/520x0/filters:focal(0x0:2936x1198):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 520w, https://cdn.vox-cdn.com/thumbor/pSomBFbim-xwUOFP-nhbagaQwvg=/0x0:2936x1198/720x0/filters:focal(0x0:2936x1198):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 720w, https://cdn.vox-cdn.com/thumbor/VdvynoZMeKgf3D-nw8wU8Xxs2Bs=/0x0:2936x1198/920x0/filters:focal(0x0:2936x1198):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 920w, https://cdn.vox-cdn.com/thumbor/tljH1USDVnD5OPgd1w4bCTZWS5g=/0x0:2936x1198/1120x0/filters:focal(0x0:2936x1198):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 1120w, https://cdn.vox-cdn.com/thumbor/oVMz6CTk7ZUotOsIIJL5kA7AxC4=/0x0:2936x1198/1320x0/filters:focal(0x0:2936x1198):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 1320w, https://cdn.vox-cdn.com/thumbor/59Gqpixe35gRVuVbdOqr2hoEcr8=/0x0:2936x1198/1520x0/filters:focal(0x0:2936x1198):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 1520w, https://cdn.vox-cdn.com/thumbor/79pquPCtkJJGzyI1-0eVX6lNIIo=/0x0:2936x1198/1720x0/filters:focal(0x0:2936x1198):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 1720w, https://cdn.vox-cdn.com/thumbor/W7SnniMVZCWiPWqvAD7H4svF4TE=/0x0:2936x1198/1920x0/filters:focal(0x0:2936x1198):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387183/Model2_a_flight_attendant.png 1920w" /></picture>
<figcaption>Courtesy of OpenAI</figcaption>
</figure>
OpenAI is well aware that DALL-E 2 generates results exhibiting gender and racial bias. In fact, the examples above are from the company’s own <a href="https://link.vox.com/click/27344967.18134/aHR0cHM6Ly9naXRodWIuY29tL29wZW5haS9kYWxsZS0yLXByZXZpZXcvYmxvYi9tYWluL3N5c3RlbS1jYXJkLm1k/608adc2191954c3cef02cd73Dbb1d1b91">“Risks and Limitations” document</a>, which you’ll find if you scroll to the bottom of the main <a href="https://link.vox.com/click/27344967.18134/aHR0cHM6Ly9vcGVuYWkuY29tL2RhbGwtZS0yLw/608adc2191954c3cef02cd73Cfcf1e0d0">DALL-E 2 webpage</a>.
OpenAI researchers made some attempts to resolve bias and fairness problems. But they couldn’t really root out these problems in an effective way because different solutions result in different trade-offs.
For example, the researchers wanted to filter out sexual content from the training data because that could lead to disproportionate harm to women. But they found that when they tried to filter that out, DALL-E 2 generated fewer images of women in general. That’s no good, because it leads to another kind of harm to women: erasure.
OpenAI is far from the only artificial intelligence company dealing with bias problems and trade-offs. It’s a challenge for the entire AI community.
“Bias is a huge industry-wide problem that no one has a great, foolproof answer to,” Miles Brundage, the head of policy research at OpenAI, told me. “So a lot of the work right now is just being transparent and upfront with users about the remaining limitations.”
<h3>Why release a biased AI model?</h3>
In February, before DALL-E 2 was released, OpenAI invited 23 external researchers to “red team” it — engineering-speak for trying to find as many flaws and vulnerabilities in it as possible, so the system could be improved. One of the main suggestions the red team made was to limit the initial release to only trusted users.
To its credit, OpenAI adopted this suggestion. For now, only about 400 people (a mix of OpenAI’s employees and board members, plus hand-picked academics and creatives) get to use DALL-E 2, and only for non-commercial purposes.
That’s a change from how OpenAI chose to deploy GPT-3, a text generator hailed for its potential to enhance our creativity. Given a phrase or two written by a human, it can add on more phrases that sound uncannily human-like. But it’s shown <a href="https://link.vox.com/click/27344967.18134/aHR0cHM6Ly93d3cudm94LmNvbS9mdXR1cmUtcGVyZmVjdC8yMjY3MjQxNC9haS1hcnRpZmljaWFsLWludGVsbGlnZW5jZS1ncHQtMy1iaWFzLW11c2xpbQ/608adc2191954c3cef02cd73Cdee78c36">bias against certain groups, like Muslims,</a> whom it disproportionately associates with violence and terrorism. OpenAI knew about the bias problems but released the model anyway to a limited group of vetted developers and companies, who could use GPT-3 for commercial purposes.
Last year, I asked Sandhini Agarwal, a researcher on OpenAI’s policy team, whether it makes sense that GPT-3 was being probed for bias by scholars even as it was released to some commercial actors. She <a href="https://link.vox.com/click/27344967.18134/aHR0cHM6Ly93d3cudm94LmNvbS9mdXR1cmUtcGVyZmVjdC8yMjY3MjQxNC9haS1hcnRpZmljaWFsLWludGVsbGlnZW5jZS1ncHQtMy1iaWFzLW11c2xpbQ/608adc2191954c3cef02cd73Ddee78c36">said</a> that going forward, “That’s a good thing for us to think about. You’re right that, so far, our strategy has been to have it happen in parallel. And maybe that should change for future models.”
The fact that the deployment approach has changed for DALL-E 2 seems like a positive step. Yet, as DALL-E 2’s “Risks and Limitations” document acknowledges, “even if the Preview itself is not directly harmful, its demonstration of the potential of this technology could motivate various actors to increase their investment in related technologies and tactics.”
And you’ve got to wonder: Is that acceleration a good thing, at this stage? Do we really want to be building and launching these models now, knowing it can spur others to release their versions even quicker?
Some experts <a href="https://link.vox.com/click/27344967.18134/aHR0cHM6Ly90d2l0dGVyLmNvbS9DYXRoZXJpbmVGbGljay9zdGF0dXMvMTUxMjM5NzAxOTUxNTY5NTEwNw/608adc2191954c3cef02cd73B0ed52d09">argue</a> that since we know there are problems with the models and we don’t know how to solve them, we should give AI ethics research time to catch up to the advances and address some of the problems, before continuing to build and release new tech.
Helen Ngo, an affiliated researcher with the Stanford Institute for Human-Centered AI, says one thing we desperately need is standard metrics for bias. <a href="https://link.vox.com/click/27344967.18134/aHR0cHM6Ly9haWluZGV4LnN0YW5mb3JkLmVkdS93cC1jb250ZW50L3VwbG9hZHMvMjAyMi8wMy8yMDIyLUFJLUluZGV4LVJlcG9ydF9DaGFwdGVyLTMucGRm/608adc2191954c3cef02cd73Bc1cef69b">A bit of work has been done</a> on measuring, say, how likely certain attributes are to be associated with certain groups. “But it’s super understudied,” Ngo said. “We haven’t really put together industry standards or norms yet on how to go about measuring these issues” — never mind solving them.
OpenAI’s Brundage told me that letting a limited group of users play around with an AI model allows researchers to learn more about the issues that would crop up in the real world. “There’s a lot you can’t predict, so it’s valuable to get in contact with reality,” he said.
That’s true enough, but since we already know about many of the problems that repeatedly arise in AI, it’s not clear that this is a strong enough justification for launching the model now, even in a limited way.
<h3>The problem of misaligned incentives in the AI industry</h3>
Brundage also noted another motivation at OpenAI: competition. “Some of the researchers internally were excited to get this out in the world because they were seeing that others were catching up,” he said.
That spirit of competition is a natural impulse for anyone involved in creating transformative tech. It’s also to be expected in any organization that aims to make a profit. Being first out of the gate is rewarded, and those who finish second are rarely remembered in Silicon Valley.
As the team at Anthropic, an AI safety and research company, put it in a <a href="https://arxiv.org/pdf/2202.07785.pdf">recent paper</a>, “The economic incentives to build such models, and the prestige incentives to announce them, are quite strong.”
But it’s easy to see how these incentives may be misaligned for producing AI that truly benefits all of humanity. Rather than assuming that other actors will inevitably create and deploy these models, so there’s no point in holding off, we should ask the question: How can we actually change the underlying incentive structure that drives all actors?
The Anthropic team offers several ideas. One of their observations is that over the past few years, a lot of the splashiest AI research has been migrating from academia to industry. To run large-scale AI experiments these days, you need a ton of computing power — more than 300,000 times what you needed a decade ago — as well as top technical talent. That’s both expensive and scarce, and the resulting cost is often prohibitive in an academic setting.
So one solution would be to give more resources to academic researchers; since they don’t have a profit incentive to commercially deploy their models quickly the same way industry researchers do, they can serve as a counterweight. Specifically, countries could develop <a href="https://hai.stanford.edu/policy/national-research-cloud#:~:text=A%20National%20Research%20Cloud%20(NRC,needed%20for%20education%20and%20research.">national research clouds</a> to give academics access to free, or at least cheap, computing power; there’s already an existing example of this in <a href="https://www.computecanada.ca/">Compute Canada</a>, which coordinates access to powerful computing resources for Canadian researchers.
The Anthropic team also recommends exploring regulation that would change the incentives. “To do this,” they write, “there will be a combination of soft regulation (e.g., the creation of voluntary best practices by industry, academia, civil society, and government), and hard regulation (e.g., transferring these best practices into standards and legislation).”
Although some good new norms have been adopted voluntarily within the AI community in recent years — like publishing “model cards,” which document a model’s risks, as OpenAI did for DALL-E 2 — the community hasn’t yet created repeatable standards that make it clear how developers should measure and mitigate those risks.
“This lack of standards makes it both more challenging to deploy systems, as developers may need to determine their own policies for deployment, and it also makes deployments inherently risky, as there’s less shared knowledge about what ‘safe’ deployments look like,” the Anthropic team writes. “We are, in a sense, building the plane as it is taking off.”
 
 
source <a href="https://www.vox.com/future-perfect/23023538/ai-dalle-2-openai-bias-gpt-3-incentives">https://www.vox.com/future-perfect/23023538/ai-dalle-2-openai-bias-gpt-3-incentives</a>

vox A new AI draws delightful and not-so-delightful images

You may have seen some weird and whimsical pictures floating around the internet recently. There’s a Shiba Inu dog wearing a beret and black...

A new Supreme Court case makes George W. Bush look like a racial justice crusader

<div><img class="ff-og-image-inserted" src="https://cdn.vox-cdn.com/thumbor/l9fhFpYF1BikrYgeLO6Q--Z0dMw=/0x32:4293x2280/fit-in/1200x630/cdn.vox-cdn.com/uploads/chorus_asset/file/23384784/534960190.jpg" /></div>
A new case just arrived on the Supreme Court’s “<a href="https://www.vox.com/2020/8/11/21356913/supreme-court-shadow-docket-jail-asylum-covid-immigrants-sonia-sotomayor-barnes-ahlman">shadow docket</a>” that could upend a quarter-century of higher education policy and end diversity programs that were once on the cutting edge of conservative thinking. <a href="https://www.supremecourt.gov/search.aspx?filename=/docket/docketfiles/html/public/21a590.html">Coalition for TJ v. Fairfax County School Board</a> is a significant escalation in the school admissions wars, because it rests on the assumption that the current crop of Republican judges will not tolerate diversity programs that do not explicitly consider race.
Twenty-five years ago, as governor of Texas, George W. Bush signed a law creating the state’s “<a href="https://news.utexas.edu/topics-in-the-news/top-10-percent-law/#:~:text=The%20%E2%80%9CTop%2010%20Percent%20Law,to%20all%20state-funded%20universities.">top 10 percent</a>” rule, which, as the name implies, guarantees Texas students who graduate in the top 10 percent of their high school class admission to state-funded universities. The program is still in effect today, although the state’s flagship school, the University of Texas at Austin, now only accepts the top 6 percent of students due to an increase in applicants.
The 10 percent rule was enacted in response to a 1996 federal appeals court decision, which <a href="https://law.justia.com/cases/federal/appellate-courts/F3/78/932/504514/">struck down an affirmative action program</a> at UT-Austin’s law school. But it quickly took on a political life of its own. As a <a href="https://www.c-span.org/video/?155911-1/bush-campaign-news-conference">candidate for president</a>, and <a href="https://www.nytimes.com/2003/01/15/politics/bushs-statement-on-affirmative-action.html">later as president</a>, Bush touted the 10 percent plan as a conservative alternative to affirmative action programs that explicitly took account of race when deciding who to admit.
The idea behind the plan was that it would open the doors of Texas’s best public universities to students at predominantly Black or Latino high schools, many of whom historically were unlikely to attend places like UT-Austin.
And yet, this program, which was a centerpiece of Bush’s higher education proposals and which has been emulated by <a href="https://www.fldoe.org/schools/k-12-public-schools/bosss/talented-twenty-program/">red</a> and <a href="https://admission.universityofcalifornia.edu/admission-requirements/freshman-requirements/california-residents/statewide-guarantee/">blue</a> states alike, is now threatened by the Coalition for TJ case pending before the Supreme Court. Coalition for TJ involves a highly selective public high school that switched less than two years ago to an admissions process that mirrors the Texas rule, partially to create a more diverse student body. The arguments advanced by the plaintiffs in this case potentially threaten any program undertaken for the purpose of fostering diversity at selective schools.
The Coalition for TJ plaintiffs, moreover, have a very real shot of prevailing. The Supreme Court, with its 6-3 Republican supermajority, is <a href="https://www.vox.com/2022/3/23/22993107/supreme-court-wisconsin-race-gerrymander-voting-rights-act-legislature-elections-commission">increasingly hostile</a> toward any effort to lift up racially disadvantaged groups. And it is widely expected to <a href="https://www.vox.com/2022/1/24/22526151/supreme-court-affirmative-action-harvard">strike down affirmative action programs</a> at Harvard and the University of North Carolina next year.
But a victory for these plaintiffs would still be an enormous escalation by the Supreme Court, as it would potentially rule out programs that are race-neutral — meaning that they do not require school officials to consider the race of individual applicants when deciding who to admit — but that were enacted in order to foster greater diversity.
<h3>“Race-conscious” versus “race-neutral” programs, briefly explained</h3>
The Supreme Court has heard a string of cases, stretching back to <a href="https://www.law.cornell.edu/supremecourt/text/438/265">Regents of the University of California v. Bakke</a> (1978), involving university admissions programs that are “race-conscious,” meaning that they explicitly took account of race when deciding which students to admit. In Bakke, for example, the Court struck down a medical school admissions program that set aside 16 of the school’s 100 seats in the incoming class for Black, Asian, Native American, or “Chicano” applicants.
Bakke and other, more recent decisions, however, also made clear that race-conscious programs are sometimes allowed. While schools cannot use quotas or <a href="https://www.law.cornell.edu/supct/html/02-516.ZO.html">other mathematical formulas</a> that give an advantage to applicants of a certain race, current law permits schools to <a href="https://www.vox.com/22301135/supreme-court-affirmative-action-harvard-college-race-students-for-fair-admission-ed-blum">consider race when deciding among similarly qualified applicants</a>, all of whom are likely to thrive at the school.
That could change soon, however. Right now, <a href="https://www.vox.com/2022/1/24/22526151/supreme-court-affirmative-action-harvard">two cases are pending before the Supreme Court</a> which challenge race-conscious admissions programs at Harvard and the University of North Carolina. These more modern affirmative action programs also give some preference to applicants who would add racial diversity to the school, but only in fairly marginal cases. Under Harvard’s system, for example, race is one of several factors — other such factors include <a href="https://www.vox.com/22301135/supreme-court-affirmative-action-harvard-college-race-students-for-fair-admission-ed-blum">athletic ability, legacy status, or economic disadvantage</a> — which can “tip” an outstanding applicant who would otherwise be rejected into the small pool of Harvard applicants who are accepted.
Coalition for TJ, by contrast, involves a race-neutral system for selecting who is admitted to the <a href="https://tjhsst.fcps.edu/">Thomas Jefferson High School for Science and Technology</a> (“TJ”), a public magnet school in northern Virginia known for its outstanding STEM instruction and high levels of college matriculation for graduates. TJ draws its students from nearby middle schools.
Until recently, TJ used a <a href="https://www.supremecourt.gov/DocketPDF/21/21A590/220615/20220408135204693_APPENDIX%20to%20April%202022%20Appl%20to%20USSC.pdf">battery of three standardized tests</a> to screen applicants. The top performers on these tests were then designated “semifinalists,” and admitted students were selected from among these semifinalists based on their GPA, test scores, teacher recommendations, and several writing assignments that semifinalists submitted as part of the application process.
In late 2020, the school <a href="https://www.supremecourt.gov/DocketPDF/21/21A590/220615/20220408135204693_APPENDIX%20to%20April%202022%20Appl%20to%20USSC.pdf">changed its admissions process</a> and modeled it after Texas’s 10 percent plan. Under the new system, each middle school that is eligible to send students to TJ receives a certain number of slots equivalent to 1.5 percent of the school’s eighth grade class size — so, much like the Texas system, the top applicants from each of these middle schools are guaranteed admission.
After these seats are filled, the school also admits about 100 more students, who are evaluated based on factors such as their GPA and an essay submitted by applicants. Certain underrepresented applicants, such as those from middle schools that historically sent few students to TJ, or students from low-income families, are also given a preference.
Notably, neither the old system nor the new system takes explicit account of an applicant’s race. Indeed, under the new system, school officials who screen applicants are <a href="https://www.supremecourt.gov/DocketPDF/21/21A590/220615/20220408135204693_APPENDIX%20to%20April%202022%20Appl%20to%20USSC.pdf">not told each applicant’s race, gender, or name</a>. TJ’s admissions program, in other words, is “race-neutral,” using a set of criteria to screen applicants that do not include race.
Yet, while TJ’s system is race-neutral, there is considerable evidence that the local school board decided to change the school’s admissions process, at least in part, because of concerns that the old process did not produce a student body that resembled the racial demographics of the local community.
In June of 2020, for example, the school released demographic data on its freshman class which, according to a federal judge, showed that the “<a href="https://www.supremecourt.gov/DocketPDF/21/21A590/220615/20220408135204693_APPENDIX%20to%20April%202022%20Appl%20to%20USSC.pdf">number of Black students admitted was too small to report</a>.” Shortly thereafter, the school principal wrote a message to students and parents stating that “we each have a responsibility to our community to speak up and take actions that counter racism and discrimination in our society,” and noting that the school’s student body “did not reflect the racial composition” in the local school district. A school board member later commented that “we must recognize the ... <a href="https://www.supremecourt.gov/DocketPDF/21/21A590/220615/20220408135204693_APPENDIX%20to%20April%202022%20Appl%20to%20USSC.pdf">unacceptable numbers of African Americans that have been accepted to TJ</a>.”
In the final five years that the school used its old admissions process, Asian Americans received at least 65 percent of offers to matriculate at TJ. In the first year that the school used the new process, by contrast, just over 54 percent of admitted students were Asian American. Meanwhile, other racial groups gained ground. About 7 percent of the students offered admission in 2021 were Black.
<h3>The Supreme Court has historically favored race-neutral programs that increase racial diversity</h3>
The Coalition for TJ plaintiffs emphasize the evidence that TJ changed its admissions process as part of an intentional effort to racially diversify the school. According to their <a href="https://www.supremecourt.gov/DocketPDF/21/21A590/220615/20220408135038911_FINAL%20FINAL%20TJ%20APPLICATION.pdf">court filing</a> asking the justices to block the new system, this system requires Asian American applicants “to compete for seats at TJ in a system intended to discriminate against them because of their race.”
If this argument — that race-neutral programs are suspect if they are motivated by a desire to increase racial diversity — succeeds, it could sound the death knell for similar race-neutral programs intended to diversify schools and universities, including the top 10 percent plan favored by Bush. Advocates of programs like it haven’t been shy about their own intentions to use race-neutral means to foster student diversity.
As a candidate for president, for example, Bush touted Texas’s program and a similar program in Florida as a way to “<a href="https://www.c-span.org/video/?155911-1/bush-campaign-news-conference">affect the pool of applicants of minority students available for higher ed in a positive way</a>.” Similarly, in a 2003 speech delivered when he was president, Bush praised top 10 percent-style programs in California, Florida, and Texas as a way that “<a href="https://www.nytimes.com/2003/01/15/politics/bushs-statement-on-affirmative-action.html">diversity can be achieved without using quotas</a>.”
The Supreme Court’s affirmative action decisions, moreover, largely endorse Bush’s view that schools may try to achieve greater levels of racial diversity, and that race-neutral programs are the preferred method to do so — although the Court’s decisions also give schools slightly more leeway to use race-conscious programs than Bush would have allowed.
In <a href="https://www.law.cornell.edu/supct/html/02-241.ZO.html">Grutter v. Bollinger</a> (2003), for example, the Court acknowledged that “numerous studies show that student body diversity promotes learning outcomes, and ‘better prepares students for an increasingly diverse workforce and society, and better prepares them as professionals.’” The idea is that students don’t just learn from their instructors, they also learn from the varied experiences of their classmates. So a student of any race who attends a racially diverse school will receive a superior education.
More recently, in <a href="https://scholar.google.com/scholar_case?case=4733326500100664777&hl=en&as_sdt=6&as_vis=1&oi=scholarr">Fisher v. University of Texas at Austin</a> (2016), the Court echoed the idea that the desire to increase “‘student body diversity’ ... is, in substantial measure, an academic judgment to which some, but not complete, judicial deference is proper.” While racial quotas and the like are forbidden, schools have some leeway to set admissions standards that foster diversity.
Fisher also held that race-neutral methods of promoting diversity are preferred to race-conscious methods. Indeed, if a school wishes to use race-conscious admissions standards, it must first prove that a race-neutral method “would not promote its interest in the educational benefits of diversity ‘about as well and at tolerable administrative expense.’”
Under current law, in other words, public schools and universities have a legitimate interest in fostering racial diversity, and they may intentionally design their admissions standards to increase the likelihood that students from underrepresented racial groups are admitted. Schools with race-conscious admissions programs may struggle to justify those programs in court, but the Supreme Court has historically treated race-neutral programs intended to enhance diversity as benign.
But there’s no guarantee that the Court will continue to view such race-neutral programs as acceptable. Fisher was a 4-3 decision, with retired Justice Anthony Kennedy writing the majority opinion, and the late Justice Ruth Bader Ginsburg joining the majority. Both Kennedy and Ginsburg were replaced by archconservative Trump appointees. (The reason only seven justices decided Fisher is that the case was handed down after Justice Antonin Scalia’s death created a vacancy on the Court, and Justice Elena Kagan was recused.)
The Court’s current Republican supermajority has shown extraordinary hostility toward laws intended to promote racial equality, and it is well to the right of an earlier generation of Republicans, like former President Bush. In 2006, for example, Bush signed legislation <a href="https://www.nbcnews.com/id/wbna14059113">reauthorizing the Voting Rights Act</a>, which forbids race discrimination in elections, but the current Supreme Court has since <a href="https://www.vox.com/2022/3/23/22993107/supreme-court-wisconsin-race-gerrymander-voting-rights-act-legislature-elections-commission">largely dismantled this historic piece of civil rights legislation</a>.
It’s not hard to imagine, in other words, that the Court’s current majority could hold that any policy motivated by a desire to increase opportunities for underrepresented racial minorities is constitutionally suspect.
 
 
source <a href="https://www.vox.com/2022/4/14/23022265/supreme-court-affirmative-action-coalition-tj-fairfax-school-george-bush-race">https://www.vox.com/2022/4/14/23022265/supreme-court-affirmative-action-coalition-tj-fairfax-school-george-bush-race</a>

vox A new Supreme Court case makes George W. Bush look like a racial justice crusader

A new case just arrived on the Supreme Court’s “ shadow docket ” that could upend a quarter-century of higher education policy and end dive...

How Democrats learned to stop worrying and love the gerrymander

The grinding battle over congressional redistricting is drawing to a close. And, contrary to <a href="https://www.vox.com/22632427/redistricting-gerrymandering-house-republicans">expectations</a> that the process would result in big Republican gains, the final House of Representatives map may well improve somewhat for Democrats.
The main reason is gerrymandering — redrawing of district lines for partisan benefit. Republicans built on their existing gerrymanders to try to expand their House advantage, but Democrats fired back even more powerfully with gerrymanders of their own.
Basically, Democrats saved themselves by resorting to a tactic they’ve previously denounced as not only unfair but downright unethical — House Speaker Nancy Pelosi <a href="https://www.speaker.gov/newsroom/62719-3">called</a> gerrymandering “unjust and deeply dangerous” in 2019. But in the absence of national reforms banning the practice, refusing to gerrymander would have meant effective unilateral disarmament, ceding the GOP a significant advantage in the battle for control over the House.
Redistricting has proceeded like a tug of war. As state legislatures, judges, and commissions have approved new maps, creating more safe or swing districts in various states, the underlying partisanship of the median House district has been pulled in one direction, and then the other. The most powerful pulls came from either state legislatures that gerrymandered, or state courts that struck down certain gerrymandered maps, as this graphic shows:
<figure class="e-image"><picture class="c-picture" data-cid="site/picture_element-1649966741_3870_80578" data-cdata='{"asset_id":23387879,"ratio":"*"}'><source srcset="https://cdn.vox-cdn.com/thumbor/B6k1M_kF_vRiIKpYhTR8W7nvpVo=/0x0:1080x1641/320x0/filters:focal(0x0:1080x1641):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 320w, https://cdn.vox-cdn.com/thumbor/I3YWQfw5EJPqb_YYsgVGPH2KO5U=/0x0:1080x1641/520x0/filters:focal(0x0:1080x1641):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 520w, https://cdn.vox-cdn.com/thumbor/ZwVG5ksyxuOxCwpYTcy_a1JMif8=/0x0:1080x1641/720x0/filters:focal(0x0:1080x1641):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 720w, https://cdn.vox-cdn.com/thumbor/wytWhFHHq-rXq40KHAzZPBigr8k=/0x0:1080x1641/920x0/filters:focal(0x0:1080x1641):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 920w, https://cdn.vox-cdn.com/thumbor/-y68m5EFoQdBg49PQiIWisXzHvc=/0x0:1080x1641/1120x0/filters:focal(0x0:1080x1641):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 1120w, https://cdn.vox-cdn.com/thumbor/3BPmaZd0EsU315RkQcXt-ljTlCs=/0x0:1080x1641/1320x0/filters:focal(0x0:1080x1641):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 1320w, https://cdn.vox-cdn.com/thumbor/I1Uu7oCVEFBzo-CQMoJELaBM0qc=/0x0:1080x1641/1520x0/filters:focal(0x0:1080x1641):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 1520w, https://cdn.vox-cdn.com/thumbor/e1GV_epC7gwDQjr5Bj2hobqnoG0=/0x0:1080x1641/1720x0/filters:focal(0x0:1080x1641):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 1720w, https://cdn.vox-cdn.com/thumbor/chqhEJp1mVBXHHxPhQ4GbJW9BPg=/0x0:1080x1641/1920x0/filters:focal(0x0:1080x1641):format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 1920w" sizes="(min-width: 1221px) 846px, (min-width: 880px) calc(100vw - 334px), 100vw" type="image/webp" /> <img alt="" data-upload-width="1080" sizes="(min-width: 1221px) 846px, (min-width: 880px) calc(100vw - 334px), 100vw" src="https://cdn.vox-cdn.com/thumbor/dwdRkMf4VwXQ5PF6Dyr1But0qDk=/0x0:1080x1641/1200x0/filters:focal(0x0:1080x1641):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png" srcset="https://cdn.vox-cdn.com/thumbor/RJPQ-5Efo81gFuVrgJTdTOnAilY=/0x0:1080x1641/320x0/filters:focal(0x0:1080x1641):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 320w, https://cdn.vox-cdn.com/thumbor/qycZt4NazgoXbz1Gpzbi_guPgB8=/0x0:1080x1641/520x0/filters:focal(0x0:1080x1641):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 520w, https://cdn.vox-cdn.com/thumbor/-nE8vg3agI1xRGyx4wJI5hETaqE=/0x0:1080x1641/720x0/filters:focal(0x0:1080x1641):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 720w, https://cdn.vox-cdn.com/thumbor/6m9wep22mvAwOoEcm6rXkeDnTKY=/0x0:1080x1641/920x0/filters:focal(0x0:1080x1641):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 920w, https://cdn.vox-cdn.com/thumbor/NpZs1WJaDjDkga_Zd-XElsti2BM=/0x0:1080x1641/1120x0/filters:focal(0x0:1080x1641):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 1120w, https://cdn.vox-cdn.com/thumbor/tvk8BOs59G8qdAkEYOwYfF-0ZWc=/0x0:1080x1641/1320x0/filters:focal(0x0:1080x1641):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 1320w, https://cdn.vox-cdn.com/thumbor/WGgST-iQ5tRSLyonqUziQn0HFk0=/0x0:1080x1641/1520x0/filters:focal(0x0:1080x1641):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 1520w, https://cdn.vox-cdn.com/thumbor/9ahLMBotgJDKJ2OKBzULiY2l8to=/0x0:1080x1641/1720x0/filters:focal(0x0:1080x1641):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 1720w, https://cdn.vox-cdn.com/thumbor/JfjZimMgm3l6t6I2m5izxotPM3E=/0x0:1080x1641/1920x0/filters:focal(0x0:1080x1641):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/23387879/redistrict_final___1_.png 1920w" /></picture> <cite>Amanda Northrop/Vox</cite></figure>
This cycle’s Republican gerrymanders pulled the median district (which already leaned 2 percentage points to the right) another point further right. But state court rulings striking down North Carolina and Ohio maps effectively wiped out most of that net gain.
Meanwhile, Democratic gerrymanders in states like New York and Illinois pulled the median district nearly 3 points leftward, so it was actually close to neutral. (Joe Biden’s margin in the median district would have nearly matched his national popular vote margin in the 2020 presidential election.) But an <a href="https://www.npr.org/2022/04/12/1092414662/gov-desantis-takes-over-congressional-redistricting-in-florida">aggressive gerrymander</a> in GOP-controlled Florida could soon shift things right again, if approved. Other state court rulings could shift things further, <a href="https://www.nytimes.com/2022/04/04/nyregion/redistricting-stay-new-york.html">particularly in New York</a>, where Democrats’ gerrymander is under scrutiny.
Currently, it looks like there will be close to an equal number of districts leaning left and right of the national average, with a slight edge to Republicans in the median district.
Now, it’s entirely possible, perhaps likely, that Democrats will still lose badly in House elections this fall — the party has a small majority, President Biden is unpopular, and the historical pattern is for the incumbent’s party <a href="https://www.vox.com/22899204/midterm-elections-president-biden-thermostatic-opinion">to struggle in the midterms</a>. But unlike much of the previous decade, the underlying map may be at least somewhat less biased in Republicans’ favor.
<h3>How Republicans won a big gerrymandering advantage a decade ago, then saw it shrink</h3>
The last national redistricting happened after Republicans won sweeping victories in the 2010 midterms, giving them control over many state legislatures and governorships. They used that power to draw lines that gave them a big advantage in the House.
By 2012, when that last redistricting was finished, the median House district <a href="https://www.nytimes.com/2022/03/10/us/politics/redistricting-congressional-maps-elections.html">leaned nearly 6 percentage points</a> further toward Republicans presidentially than that year’s national popular vote. The results were clear: Obama won nationally by about 4 points in 2012, but he lost the median district by about 2 points. What’s more, 55 percent of the overall House districts (240 out of 435) leaned Republican, <a href="https://www.nytimes.com/2022/03/10/us/politics/redistricting-congressional-maps-elections.html">per the New York Times</a>. That sizable advantage helped Republicans hold the House in 2012 despite Obama’s national win.
But over the course of the ensuing decade, that GOP advantage significantly eroded. Changes in demographic voting patterns made many suburban districts less safe for Republicans. Meanwhile, courts struck down Republican gerrymanders in states like <a href="https://www.vox.com/2015/12/5/9851152/florida-gerrymandering-ruling">Florida</a>, <a href="https://www.vox.com/2018/1/22/16920636/pennsylvania-gerrymander-ruling-house">Pennsylvania</a>, and <a href="https://www.vox.com/policy-and-politics/2015/6/5/8737667/virginia-racial-gerrymandering-decision">Virginia</a>. By the time the 2020 results were in, the median House district still leaned toward Republicans, but only by 2 points, rather than 6 points. And about 52 percent of districts (228 out of 435) had a Republican lean.
That was real progress for Democrats on reducing the bias of the House, but it was accompanied by disappointment. First, though Democrats performed well in the 2018 and 2020 elections, they fell short of retaking several key governorships and legislatures, meaning Republicans would have the power to gerrymander again in these states. Second, the party had hoped the Supreme Court would declare partisan gerrymandering unconstitutional, banning it nationwide, but Trump’s appointees moved to the court to the right and the conservative majority <a href="https://www.npr.org/2019/06/27/731847977/supreme-court-rules-partisan-gerrymandering-is-beyond-the-reach-of-federal-court">ruled otherwise</a>. Third, efforts to pass a nationwide gerrymandering ban through the Democrat-controlled Congress under Biden <a href="https://www.vox.com/22876643/democrats-voting-rights-senate-filibuster-manchin">were stymied</a> by the Senate filibuster.
So as 2021 began, Republicans had the power of line-drawing in several swing states, as well as red states where they hadn’t yet maxed out their advantage. The GOP still had an advantage in the House map, and now it <a href="https://www.vox.com/22632427/redistricting-gerrymandering-house-republicans">seemed they</a> could entrench and expand it.
<h3>How to measure a district’s partisanship</h3>
There are varying ways to estimate the underlying partisanship of a district or an overall map, but for now, I’ll focus mainly on a simple one: how the district voted in the most recent presidential race, compared to the national popular vote. (The New York Times’s Nate Cohn <a href="https://www.nytimes.com/2022/03/10/us/politics/redistricting-congressional-maps-elections.html">used this metric</a> in his own recent analysis.)
In 2020, Biden won the popular vote by a margin of 4.4 percentage points. If he won by more than that in a given district, I’m calling that a Democratic-leaning district. If he won by less than that, or lost the district, I’m calling it a Republican-leaning district. This metric lets us look at the partisan lean for the median House district (the one necessary to give a party a majority), and also measure how many districts lean toward Democrats or Republicans overall.
Focusing on the presidential numbers won’t be a perfect guide to House results. House candidates run with their own strengths and weaknesses, and some manage to defy their districts’ underlying partisan lean. But there have been fewer such candidates lately — in 2020, only 16 out of 435 House victors won a district where the opposite party’s presidential candidate also won.
Other analysts may have slightly different specific calculations for the map’s overall lean. For this cycle, I’ve used the <a href="https://www.cookpolitical.com/">Cook Political Report</a> (an invaluable resource for anyone closely following elections), which calculated the presidential results in each new district. As an alternative, Cook also uses a metric called the <a href="https://www.cookpolitical.com/pvi-0">PVI (Partisan Voting Index)</a>, which incorporates the past two presidential elections. FiveThirtyEight has its own “<a href="https://fivethirtyeight.com/features/how-red-or-blue-is-your-state-your-congressional-district/">partisan lean</a>” score. The <a href="https://gelliottmorris.substack.com/p/new-us-house-maps-are-fairer?s=r">Economist’s G. Elliott Morris</a> argues it can be most predictive to look at the presidential election prior to the most recent one. Still, these different estimates will probably be roughly similar overall.
<h3>The Republican gerrymanders of 2021-2022</h3>
Let’s start by looking at how Republican gerrymandering attempts fared this cycle. The GOP did indeed try to expand their advantage in key states, but their overall impact was hampered by a few factors.
<ul>
<li>North Carolina was most painful for Republicans, as the state supreme court struck down their sweeping gerrymander and replaced it with a relatively balanced plan.</li>
<li>An unfavorable court ruling in Ohio also knocked down another sweeping GOP gerrymander, but the party came back with an only slightly less gerrymandered map that the court will apparently allow to be put in place this year.</li>
<li>In Texas, where the preexisting GOP gerrymander was weakening due to changes in voting patterns such as suburbs trending Democratic, the GOP made a strategic decision to focus on maximizing safety for incumbents rather than expanding their reach. So they wiped out most swing districts in the state and Republican incumbents will be harder for Democrats to defeat, but the number of districts that voted more for Trump than the national average in 2020 is staying the same (25 out of now 38 in the state, a major Republican advantage in a state Trump only won by 5.6 percentage points).</li>
</ul>
In the finalized maps so far, then, Republicans have ended up with just a handful of new districts leaning in their favor. But that’s compared to a map that was already favorable to them, and they managed to preserve or strengthen preexisting pro-GOP maps in key states.
However, one other state may soon give them a big assist: Florida Gov. Ron DeSantis (R) and GOP state legislative leaders have been at odds for months on just how much the House map should tilt in Republicans’ favor, with DeSantis pushing for a more extreme gerrymander. And just this week, DeSantis <a href="https://www.npr.org/2022/04/12/1092414662/gov-desantis-takes-over-congressional-redistricting-in-florida">appeared to win</a>: The legislature said it would approve whatever maps he wanted. Florida alone could move the median district’s margin one percentage point to the right.
<h3>The Democratic gerrymanders of 2021-2022</h3>
Democrats, meanwhile, really went to town with gerrymanders of their own in states they controlled:
<ul>
<li>In New York and Illinois, Democrats made big gains. In each state, they erased four Republican-leaning districts and created three Democratic-leaning ones (each state lost a seat due to population decline). New York’s new map will have 22 of 26 districts leaning to the left of the national average, and Illinois will have 14 of 17 tilted to the left. One potential hiccup for Democrats is that a lower court judge struck down the New York map, but so far the state’s high court has stayed that decision, and it’s not clear if it <a href="https://www.nytimes.com/2022/04/08/nyregion/redistricting-maps-new-york.html">will go into effect</a>.</li>
<li>In Oregon, which is gaining a seat due to population growth, Democrats drew a map with five Democratic-leaning districts and one Republican-leaning district — an improvement for them over the 3-2 status quo.</li>
<li>In Nevada, Democrats turned a map where three of four districts leaned Republican to one where three leaned Democratic.</li>
<li>In New Mexico, which had two Democratic-leaning districts and one GOP-leaning one, Democrats wiped away that Republican district in an attempt to create a 3-0 map.</li>
</ul>
That amounts to wiping out 12 Republican districts and creating 11 Democratic districts — an enormous impact on the overall map.
One caveat is that some analysts think Democrats may have spread themselves a bit too thin in some of these maps by creating several districts that lean Democratic, but not strongly so, such that Republicans could very plausibly win in these areas in a strong GOP year. This is the case particularly in Illinois, Nevada, and New Mexico. Still, in a Republican wave year, the GOP is quite likely to win control of the House regardless of what happens in these states. But the lean-Democratic districts tip the balance when there’s a close national contest.
There were also states in which redistricting was handled by commissions (rather than state legislatures), or where power was divided. For overall partisan balance, these proved to be close to a wash — for instance, commissions eliminated a Republican-leaning district in both California and Michigan, but created a Republican-leaning district in both Arizona and Colorado. (Some Democrats are rueing the lost opportunities to gerrymander Colorado and Virginia, states where they had full control in 2021, because redistricting authority had been given to commissions there.)
Overall, then, the 2022 redistricting wars turned out to come down to a battle of the gerrymanders — and Democrats’ ended up being more impactful.
<h3>If Democrats had not gerrymandered, they’d face a much more Republican-leaning map</h3>
Democrats have spent the past decade deriding gerrymandering as unethical and immoral, and trying to get it banned across the country.
Yet the plain reality is that, if they had decided not to do any of it, Republicans would not only have retained their existing advantage in the House map, they would have expanded it.
Though some states haven’t finalized their maps yet and these numbers can change, it’s currently looking like around 218 districts will have voted more for Trump than the national average in 2020, and 217 districts will have voted more for Biden (per the Cook Political Report’s numbers). Furthermore, Biden’s margin of victory in the median district would be about 1 percentage point lower than his margin of victory nationally. That’s not perfectly balanced, but it’s pretty balanced — meaning the map itself will likely only swing outcomes in the very closest of elections.
Contrast this to a scenario where Democrats agreed to unilaterally disarm and do no gerrymandering — or where the blue states tied their own hands by adopting serious anti-gerrymandering reforms.
Assuming something close to the 2020 maps remained in these states, around 230 of the overall new districts would have voted more for Trump than the national average, and the median district would have leaned nearly 4 points to the right of the national presidential popular vote.
A similar dynamic has arisen with other good-government reform issues, like campaign finance. Democrats spent a decade condemning conservative big money and dark money, and trying without success to rein in their influence. But the party thought it would be foolish to take the high ground by forswearing those practices. And eventually, by 2020, they arguably <a href="https://www.nytimes.com/2022/01/29/us/politics/democrats-dark-money-donors.html">ended up</a> mastering them more expertly than Republicans.
Republicans believe Democrats’ appeals to ethics were always situational. They point out that Democrats only began to complain about gerrymandering so loudly once Republicans got the chance to do so much of it in 2010, and that Democratic state parties have often been eager to gerrymander when they’ve had the power to do so.
Still, all this does get at the difficulty of making reforms stick without a national solution. There’s a prisoner’s dilemma aspect to gerrymandering, in which agreeing not to get your hands dirty may well just mean agreeing to lose.
For Democrats genuinely concerned about good-government reforms, that poses a challenge. Without a national solution, is it worth it to try to keep reforming gerrymandering in blue-leaning states?
Or, if you do so, are you just a sucker?
 
 
source <a href="https://www.vox.com/22961590/redistricting-gerrymandering-house-2022-midterms">https://www.vox.com/22961590/redistricting-gerrymandering-house-2022-midterms</a>

vox How Democrats learned to stop worrying and love the gerrymander

The grinding battle over congressional redistricting is drawing to a close. And, contrary to expectations that the process would result in ...