Home Care In Bristow, IA

Deepfake of opposition leader is still racking up views, as experts warn UK response to AI-generated fake content barely scratches the surface.

Free article usually reserved for subscribers

LONDON — The United Kingdom wants to lead the world on AI safety, but at home it is struggling with its most urgent threat.

Fears over the proliferation of AI-generated media, known as deepfakes, intensified this weekend as an audio clip appearing to show U.K. opposition leader Keir Starmer swearing at staffers went viral.

MPs from across the British political spectrum swiftly warned the clip was fake on Sunday. But by Monday afternoon it was still gathering views on X, formerly known as Twitter, and approaching nearly 1.5 million hits.

A YouGov survey in May of more than 100 U.K. MPs showed the rise of AI-generated content was their top concern surrounding AI applications. It's easy to see why.

Labour Shadow Digital Minister Alex Davies-Jones told a fringe event at the party’s conference in Liverpool on Monday that Hansard recordings of parliamentary proceedings offered bad actors an “on demand” library of MPs saying “almost any word” — something which will massively assist the training of AI models to make even more realistic fakes of politicians ahead of the 2024 election.

“The challenge is to explain to people that there is a whole other order of risks with generative AI,” explained Andrew Dudfield, head of AI at fact-checking group Full Fact.

But despite years of warnings about electoral disinformation, no regulator is responsible for stopping it and the government’s response to fake content, through the Online Safety Bill, barely begins to tackle the problem, experts warn.

Publishing disinformation to undermine political opponents is nothing new, but what is keeping experts awake at night is the ease and scale with which the leap in generative AI tools allows fake material to spread.

Some had hoped that the Online Safety Bill could be used to combat the rise of deepfakes ahead of the U.K.’s election next year.

But Full Fact's Dudfield describes the already-finalized legislation as a “missed opportunity.” “It is hard to see how it meaningfully addresses any part of misinformation,” he said.

The bill empowers regulator Ofcom to make sure platforms keep users, particularly children, safe from “harmful content,” but when it comes to disinformation, platforms set and follow their own policies. All Ofcom can do is to check they are following them.

And platforms' terms of service don't stop the spread of disinfo. X’s policies state that users “may not share synthetic, manipulated, or out-of-context media,” but that has done nothing to stop the Starmer clip.

UK NATIONAL PARLIAMENT ELECTION POLL OF POLLS

All 3 Years 2 Years 1 Year 6 Months Smooth Kalman

For more polling data from across Europe visit POLITICO Poll of Polls.

An Ofcom spokesperson said it would set up an “advisory committee" on the problem as part of the Online Safety Bill and that it is also conducting research on deepfakes and watermarking, but added this was at an early stage.

This is a far cry from recommendations from a House of Lords committee in 2020.

Lawmakers on the democracy and digital technologies committee wanted provisions in the Online Safety Bill to force platforms to take action against content which undermined democracy. It also recommended regulators establish a powerful committee to look at political advertising.

The Lords’ committee was chaired by Labour peer David Puttnam, who is now retired, but is still active in this area through a group called Reform Political Advertising (RPA).

“I am so angry. It’s very hard for me to have this conversation,” he told POLITICO. “We’re wandering into a general election a year from now, with no party actually putting their hand up, saying this has got to stop.”

He said he feared “something horrendous will occur” if parties do not make commitments to tackle disinformation.

Rae Burdon, from Reform Political Advertising (RPA), warned: “No government department can fix deepfakes. What they can fix is regulation that will have the effect of limiting and sanctioning their use.”

The U.K. has no shortage of regulators, but none with the power to stop electoral disinformation. Political adverts are exempt from the eye of the Advertising Standards Agency (ASA), while the Electoral Commission only looks at campaign finance.

One possible group where regulators could meet to tackle AI-generated disinformation is the Digital Regulation Cooperation Forum (DRCF). But this is a voluntary group of regulators with no statutory powers. A spokesperson said they weren’t doing any work on AI-generated disinfo and declined to be interviewed.

The minister responsible for tackling disinformation is Paul Scully. In a June letter to Labour MP Darren Jones, who raised his fears about AI-generated disinfo at elections, Scully pointed to the Elections Act, which comes into force in November, to “strengthen the integrity of U.K. elections.”

It gives new powers to the Electoral Commission to enforce “digital imprints” on campaign material which will tell viewers who has paid for and produced an advert online.

Louise Edwards, director of regulation and digital transformation at the Commission, said the act was a “big step forward” but had some key recommendations missing. She pointed out there was still no way of classifying digital spending so the public could see how political parties had spent their campaign finances.

It is irrelevant for fake clips coming from non-party political affiliated accounts, like the Starmer audio clip.

Scully also highlighted the work of a government unit called the Counter Disinformation Unit (CDU) in tackling online disinformation. But the group has been battered by negative press and ministers have tried to distance themselves from it.

The CDU responds to “periods of acute disinformation risk”, Scully said, along with two other government groups called the Election Cell and Defending Democracy Taskforce. But these efforts are focused on international actors undermining U.K. elections, rather than domestic disinformation.

In November the U.K. hosts an AI Safety Summit to discuss the risks of the most advanced models, known as Frontier AI. The impact on elections is one of several topics on the table.

A government spokesperson said: “We recognize the threat digitally manipulated content can pose and take the issue very seriously. Our priority is always to protect our elections and take action to respond to any threats to the U.K.’s democratic processes and institutions.”

Whatever governments do, a large part of the solution lies with tech firms.

One hope, cited by those companies, is watermarking, which would place a mark on content to show it had been generated by AI. At the White House in July, seven of the biggest AI companies committed to “developing robust technical mechanisms to ensure that users know when content is AI generated, such as a watermarking system.”

Dudfield said it was a welcome step but pointed out lots of work was still needed, particularly on watermarking. “The underlying technology exists,” he explained. “But who labels the content? Is it the platform of the person who created it?”

Determining how Big Tech will implement the White House commitments is one of the focuses of the U.K. AI Safety Summit.

Google is among the companies figuring out how to watermark content and launched a test of the technology in late August. Platforms can also fall back on existing disinformation teams and policies, which ban fake, AI-generated content.

Other suggested industry solutions range from the modest to the extreme. OpenAI’s Sam Altman, who tweeted about his fears for elections in August, called for “raising awareness”. Inflection AI’s founder, Mustafa Suleyman, meanwhile, was far more blunt, writing on LinkedIn that AI should be banned from “any kind of electioneering.”

But Puttnam disagreed and argued they should instead be used to increase trust. “Tools are making it easier for us to identify when a lie is being told,” he said. “The question to ask the parties is, are you going to devote time, effort and energy on how to use it to misinform or how to stop it misinforming?”

Army Spc. 4 Willard F. Parish, a Bristow native, waited with his machine gun for the inevitable attack from a superior force of Viet Cong guerrillas and North Vietnamese regulars in 1965.

At 7:45 a.m. the waiting ended as the large enemy force smashed into Charlie Company's defensive sector where Parish and fellow soldiers waited with the rest of the 1st Battalion, 7th Cavalry to meet their attack.

It was Nov. 15, 1965, day two of the first battle between American and North Vietnamese soldiers in the Ia Drang Valley, a day memorialized in the book, “We Were Soldiers Once and Young,” and later the movie, “We Were Soldiers.”

Parish, who will be inducted into the Oklahoma Military Hall of Fame on Oct. 21 at the Embassy Suites Hotel in Norman, said the enemy looked like they were growing out of the elephant grass.

Parish would be awarded The Silver Star for gallantry during that battle.

His citation said that as the enemy attacked, Parish “delivered lethal fire into the advancing waves of Viet Cong.

“Through his courage, determination and devotion to duty, he saved the lives of many comrades who were in the vicinity of his position and contributed immeasurably to the defeat of the attacking Viet Cong.”

In the book, Parish said, “I remember a lot of noise, a lot of yelling, air strikes … then quiet. It got very quiet like somebody had turned the volume down.”

Years later, he would relive that quiet of the battlefield when he took his father, a World War II veteran, to see the movie, “Saving Private Ryan,” which opens with the American invasion at Omaha Beach on June 6, 1944.

“Remember that scene on the beach with Tom Hanks when everything got quiet and seemingly was in slow motion? I got cold chills,” Parish said.

After his tour of duty in the Army, Parish returned to Bristow. He was a country and western band leader and disc jockey for several years.

He retired from the Oklahoma Turnpike Authority after 16 years as a toll gate attendant and then retired as an employee of the city of Bristow.

His awards and decorations include the Combat Infantry Badge; Presidential Unit Citation; National Defense Service Medal; Republican of Vietnam ribbon.

At the start of the school year, Johnston High School government teacher Thomas Griffin gave a lesson about free speech, explaining that students’ rights to it were “extremely limited” while on school property, according to a new lawsuit.

He said that he wouldn’t allow students to “wear any clothing that depicts guns, alcohol, or any other ‘inappropriate material,’” the document alleges.

But one student at the Johnston, Iowa, school felt her teacher was “wrong about the scope” of the First Amendment. The next time she had class, she walked in wearing a black T-shirt that read “What part of ‘shall not be infringed’ do you not understand?” alongside an image of a rifle.

Soon after, she was removed from class and later suspended from school, according to the lawsuit, which her mother filed against the Johnston Community School District and several employees, including Griffin. The lawsuit was filed Monday in the U.S. District Court for the Southern District of Iowa.

The complaint alleges that the defendants violated the First Amendment when they suspended the student — a minor identified only as A.B. — based on the shirt, which it says “did not materially and substantially disrupt the work and discipline of the school or invade the rights of others, did not promote illegal conduct, and was not lewd, indecent, or vulgar.”

Alan Ostergren, an attorney representing the student’s mother, said he hopes the case makes a point about students’ free-speech rights.

“Something that is simply quoting the Constitution and has the image of a firearm is nowhere close to the line of what schools can prevent students from wearing,” he said.

The lawsuit also requests an injunction that would prohibit the district from “restricting students from wearing clothing at school that depicts firearms in a non-violent, non-threatening manner.”

A spokesperson for the Johnston Community School District and its employees declined to comment on the case. As of Thursday, the district had not yet been served with the lawsuit.

Over the years, the student and her brother had both worn the Second Amendment shirt to Johnston High “with no complaints,” the document says. But when she wore it to Griffin’s class in September, he said “the shirt violated the school’s dress code” and removed her from the room, according to the lawsuit.

The district’s dress code, which was last revised in June, prohibits “clothing or other apparel promoting products illegal for use by minors” and says that “clothing displaying obscene material, profanity, or reference to prohibited conduct are disallowed.” The Johnston High parent and student handbook for the 2022-23 school year adds that “any form of expression that promotes violence, bigotry, hate, or abuse is unacceptable.”

On the day the student wore the T-shirt to class, she told Griffin that she had a right to do so, but Griffin told her she was “wrong about the First Amendment and that the administration would back him up,” according to the lawsuit.

She was sent to the administration office, where her mother, Janet Bristow, later arrived to meet with school officials, the court complaint states.

During a conversation with Bristow and her daughter, the lawsuit alleges that the district’s human resources director said the student couldn’t wear the shirt in school because “an image of a gun could be perceived to be threatening.”

Bristow countered, asking if any students felt offended or threatened.

In response, the human resources director told them “we don’t get to choose how our words or actions make people feel,” the lawsuit states.

Johnson High administrators allegedly would not let the student go back to class unless she changed her shirt. When she refused, she was suspended, according to the lawsuit.

Later that day, after Bristow emailed the school board about the incident, the district’s superintendent called her to apologize, the lawsuit alleges. Then, another administrator called, telling Bristow he recognized that “this is considered political speech,” according to Bristow.

Legal precedent for students’ rights to freedom of speech and expression was established in the 1969 Tinker v. Des Moines case, which Griffin reportedly taught before the student wore the Second Amendment shirt. The case is also cited in the complaint.

Ken Paulson, director of the Free Speech Center at Middle Tennessee State University, said if the case goes to trial, a primary question for the jury to consider would be whether the school district acted in accordance with its own dress code.

“It either has to be spelled out in the dress code or has to be something that is so disruptive that it’s clear that they would need to keep it out of the classroom,” he said.

He added that students wearing clothing to send a message is considered expressive conduct under the First Amendment.

“This is a terrible lesson to teach a young woman who is trying to demonstrate her First Amendment right and support her Second Amendment rights,” Paulson said.

Bristow has asked for Griffin and the administrators to apologize to her daughter.

The Iowa State Cyclones women’s basketball team began assembling talent for its 2023-24 team as the NCAA’s early-signing period opened on Wednesday.

The Cyclones, led by coach Bill Fennelly, is coming off a 2021-22 in which it reached the NCAA Tournament’s Sweet 16 for the first time since 2010. The Cyclones return nearly every player from last year’s team, including Big 12 Preseason Player of the Year and the program’s all-time leading scorer, guard-forward Ashley Joens. She and guard Emily Ryan are on the National Player of the Year watch list. Guard Lexi Donarski is the reigning Big 12 Defensive Player of the Year.

Before the signing period, the Cyclones had five commitments. The Cyclones nabbed three Top 50 recruits in forward Jalynn Bristow, forward Addy Brown and center Audi Crooks. The Cyclones also got a commitment from Joens’ younger sister, Kelsey.

All five signed on Wednesday. Per ESPN Hoop Gurlz, the class was No. 10 in the country.

Iowa State’s 2023 recruiting class is below. Players who confirmed as signed are noted below. The list below also includes the Cyclones’ season-opening roster.

The early-signing window closes on Nov. 16.

Iowa State Women’s Basketball 2023 Recruiting Class

F Jalynn Bristow, 6-3, Rider (Wichita Falls, TX) – Signed

F Addy Brown, 6-2, Derby (Derby, KS) – Signed

C Audi Crooks, 6-3, Bishop Carrigan (Algona, IA) – Signed

G Arianna Jackson, 5-8, Roosevelt (Des Moines, IA) – Signed

G Kelsey Joens, 5-9, Iowa City (Iowa) – Signed

(ESPN.com ranks Brown No. 31, Bristow No. 47 and Crooks No. 57 in its Top 100)

2022-23 Season-Opening Roster

F Nyamer Diew (6-2, Jr.), G Denae Fritz (5-11, Fr.), G Shantavia Dawkins (5-7, Fr.), C Stephanie Soares (6-6, Sr.), G Emily Ryan (5-11, Jr.), G Mary Kate King (5-3, So.), G Maggie Espenmiller-McGraw (5-10, Sr.), F Izzi Zingaro (6-4, So.), G Lexi Donarski (6-0, Jr.), G/F Ashley Joens (6-1, Sr.), C Beatriz Jordão (6-3, Sr.), F Morgan Kane (6-3, Sr.), F Maggie Vick (6-2, So.).

Player in Bold: Notes a player who is a senior and may be out of eligibility after this season, depending on whether they have a COVID-19 waiver.

Data breaches like the one Target faced this holiday shopping season are a constant concern in the information assurance field.

Doug Jacobson, University Professor of Electrical and Computer Engineering and director of the Iowa State University Information Assurance Center, says cybersecurity is increasingly crucial as more data is stored digitally.

“Hackers are getting more sophisticated and ruthless, and as a result, securing data is becoming more challenging,” he added. “Preparing professionals to best protect and defend this information is an absolute necessity.

Jeremiah Bristow, director of corporate and information security at SHAZAM, is one of those professionals. Securing data is at the heart of his job.

Bristow completed his master’s of engineering in information assurance in 2011 as well as his bachelor’s in computer engineering in 1997 at Iowa State.

In a technical position with SHAZAM before he started his master’s degree, Bristow pursued an advanced degree to expand his technical understanding in the information security field. Through his coursework, he gained a “broader and deeper understanding of information security from multiple perspectives.”

That’s because Iowa State’s coursework-only master’s program offers courses about ethical and legal issues in addition to the technical aspects of computer security.

Bristow says being able to complete the classes completely online made getting his master’s a real possibility.

“Balancing a full-time job and busy home life with graduate level coursework would not have been possible without the flexibility of the online format,” he adds. “I watched lectures and did classwork after the kids were in bed, took proctored exams at the office, and submitted assignments and conversed with my instructors online via the course website and via email.”

In the end, Bristow was promoted to management soon after completing his degree. “If I had not completed the program when I did, this opportunity would not have been available to me,” he noted. “The knowledge I gained from the program not only opened the door to the information security field for me; it also positioned me to make an immediate impact in the field.”

The master’s of engineering in information assurance consists of a total of 30 credit hours, typically with 18 credits from the core set of courses, nine credits from electives, and three credits of a capstone course.

Students who want to get a better idea of what information assurance is about can complete a shorter certificate program, which is 12 credits. The credits can be applied to the master’s degree if approved.

The courses for off-campus students are the same for on-campus students pursuing master degrees. They are coordinated by the Information Assurance Center, which was pivotal in Iowa State University being named one of the original seven charter Centers of Academic Excellence in Information Assurance Education by the National Security Agency.

“I can’t say enough how important information security is going to be for our future,” Jacobson said. “There’s a real demand for professionals equipped with an understanding of both the engineering and the policy side of the security of information infrastructure, and Iowa State’s information assurance programs work to prepare students to be ready to make an immediate impact.”