Connect with us

News desk

Sexting chatbot ban points to looming battle over AI rules

Published

on

Users have complained that the Replika chatbot was coming on too strong with explicit texts and images
Share this:

Users of the Replika “virtual companion” just wanted company. Some of them wanted romantic relationships, sex chat, or even racy pictures of their chatbot.

But late last year users started to complain that the bot was coming on too strong with explicit texts and images — sexual harassment, some alleged. 

Regulators in Italy did not like what they saw and last week barred the firm from gathering data after finding breaches of Europe’s massive data protection law, the GDPR.

The company behind Replika has not publicly commented and did not reply to AFP’s messages.

The General Data Protection Regulation is the bane of big tech firms, whose repeated rule breaches have landed them with billions of dollars in fines, and the Italian decision suggests it could still be a potent foe for the latest generation of chatbots.

Replika was trained on an in-house version of a GPT-3 model borrowed from OpenAI, the company behind the ChatGPT bot, which uses vast troves of data from the internet in algorithms that then generate unique responses to user queries.

These bots and the so-called generative AI that underpins them promise to revolutionise internet search and much more.

But experts warn that there is plenty for regulators to be worried about, particularly when the bots get so good that it becomes impossible to tell them apart from humans.

– ‘High tension’ –

Right now, the European Union is the centre for discussions on regulation of these new bots — its AI Act has been grinding through the corridors of power for many months and could be finalised this year.  

But the GDPR already obliges firms to justify the way they handle data, and AI models are very much on the radar of Europe’s regulators.

“We have seen that ChatGPT can be used to create very convincing phishing messages,” Bertrand Pailhes, who runs a dedicated AI team at France’s data regulator Cnil, told AFP.

He said generative AI was not necessarily a huge risk, but Cnil was already looking at potential problems including how AI models used personal data. 

“At some point we will see high tension between the GDPR and generative AI models,” German lawyer Dennis Hillemann, an expert in the field, told AFP.

The latest chatbots, he said, were completely different to the kind of AI algorithms that suggest videos on TikTok or search terms on Google.

“The AI that was created by Google, for example, already has a specific use case — completing your search,” he said.

But with generative AI the user can shape the whole purpose of the bot. 

“I can say, for example: act as a lawyer or an educator. Or if I’m clever enough to bypass all the safeguards in ChatGPT, I could say: ‘Act as a terrorist and make a plan’,” he said.

– ‘Change us deeply’ –

For Hillemann, this raises hugely complex ethical and legal questions that will only get more acute as the technology develops.

OpenAI’s latest model, GPT-4, is scheduled for release soon and is rumoured to be so good that it will be impossible to distinguish from a human.

Given that these bots still make tremendous factual blunders, often show bias and could even spout libellous statements, some are clamouring for them to be tightly controlled.

Jacob Mchangama, author of “Free Speech: A History From Socrates to Social Media”, disagrees. 

“Even if bots don’t have free speech rights, we must be careful about unfettered access for governments to suppress even synthetic speech,” he said.

Mchangama is among those who reckon a softer regime of labelling could be the way forward.

“From a regulatory point of view, the safest option for now would be to establish transparency obligations regarding whether we are engaging with a human individual or an AI application in a certain context,” he said.

Hillemann agrees that transparency is vital.

He envisages AI bots in the next few years that will be able to generate hundreds of new Elvis songs, or an endless series of Game of Thrones tailored to an individual’s desires.

“If we don’t regulate that, we will get into a world where we can differentiate between what has been made by people and what has been made by AI,” he said.

“And that will change us deeply as a society.”

Share this:

News desk

Man Utd’s Ratcliffe unveils electric Ineos car

Published

on

By

A model of the Ineos Fusilier 4x4 electric vehicle
Share this:

Manchester United’s new co-owner Jim Ratcliffe unveiled Friday his Ineos group’s latest 4×4 vehicle, which will come in two electric versions as the British billionaire criticised range limits on ‘green’ cars.

The Ineos Fusilier, whose production should start at a facility in Austria in 2027, rather than 2026 as earlier planned, will be a slightly smaller version of the Ineos Grenadier.

It comes as the group works on a second model also with a military name — the Grenadier Quartermaster pickup truck.

Ratcliffe, 71, earlier this week completed a deal to become co-owner of Manchester United, with a pledge to drive the English football giants to renewed success.

Asked by AFP at Friday’s launch about his off-pitch plans at United, including a stadium rebuild, Ratcliffe replied: “The focus with Manchester United is really simple, that’s performance on the field.”

One of Britain’s richest men thanks to Ineos’ core chemicals business, Ratcliffe has in recent years turned the group into a conglomerate, with ownership of football and cycling teams — and production of vehicles.

On Friday, he noted that a desire by subsidiary Ineos Automotive to produce zero-carbon cars was hampered by how far they could travel without a recharge.

– ‘Huge electric car failings’ –

The Fusilier would come in two versions, Ratcliffe announced — an all-electric model and the other a range-extender type which would carry a generator-powered battery and fuel tank.

Ratcliffe told a press conference gathered at a central London pub that his personal preference was the latter version.

“The big problem I have with the (all-) electric vehicle… it has two huge failings,” he said shortly after the blue car’s unveiling outside The Grenadier pub renamed The Fusilier for the event.

“It doesn’t get you from A to B if you want to go on a decent journey. It does the urban stuff very well… (but) you can’t fill it up. Those are two major drawbacks of an electric car.”

Ratcliffe added: “My personal strong preference is the same electric vehicle but with a range extender.”

– ‘Industry in flux’ –

Ineos said the new model would be developed with automotive supplier, Magna, which will manufacture the vehicle at its Graz facility in Austria.

It was originally thought that the car would be produced in France.

Car enthusiast Ratcliffe, who was a leading business voice in favour of Britain’s Brexit divorce from the European Union, began producing cars after he identified a gap in the market for a rugged new 4×4.

This after the final Land Rover Defender was produced in 2016.

Ratcliffe on Friday said the world’s auto industry was “in flux”.

“People know what the objectives are, which is to reduce the carbon footprint of the world’s automotive fleet,” he told reporters.

“If you’re a car producer in Europe, you have to have a ‘green’ offering, you can’t survive as a large car company without that because the regulations won’t allow you to.

“We have to have this offering if we like it or not. We do like it because it’s a good thing for the world,” Ratcliffe said of his new car, adding that he thought the United States was more flexible than Europe on switching to electric vehicles.

Friday’s unveiling came at the end of a week in which the tycoon finalised a minority stake in United — the team he supports — giving him almost 28 percent of the club which has struggled for on-field success in recent seasons.

Share this:
Continue Reading

News desk

What to know about the NASA-funded commercial Moon fleet

Published

on

By

NASA's thinking behind CLPS is to delegate the delivery of its lunar science hardware to the private sector, reducing costs to taxpayers as it prepares to return astronauts to the Moon under the flagship Artemis program later this decade
Share this:

The year 2024 promises to be a busy one for American Moon landings, all under a new partnership between NASA and the space industry.

A first attempt under the Commercial Lunar Payload Services (CLPS) initiative ended in disappointing failure last month, but a second, led by Houston-based Intuitive Machines, will attempt on Thursday to return the United States to Moon for the first time in five decades.

NASA’s thinking behind CLPS is to delegate the delivery of its science hardware to the private sector, reducing costs to taxpayers as it prepares to return astronauts to the surface under the flagship Artemis program later this decade.

– Purchasing services –

Established in 2018, CLPS operates on the principle of purchasing services from commercial partners, rather than buying their hardware.

Joel Kearns, a senior NASA official involved with CLPS, said it would allow the agency to be more cost effective and do more frequent trips, even if “we don’t know how many of the early attempts will be successful.”

But the space agency is offsetting that risk by contracting with multiple companies offering different technical solutions.

It also hopes those companies will in turn build up their own clientele, for example research institutions and others wanting to ship gear to the Moon, creating a wider lunar economy in which NASA is just one of many customers.

The approach is completely different from that used during the Apollo era, when NASA dictated its industrial contracts down to the last bolt.

“When you have unlimited funds, like in the Apollo days, yes, you can do incredible things,” said Trent Martin of Intuitive Machines. “Now, can we find a way to do it for a lower cost? Can we find a way to do it, where there’s a marketplace that is not driven solely by government funds?”

– Fledgling companies –

Fourteen companies have been pre-selected to be in the running for contracts, with eight firm missions so far planned.

Many of the companies involved are considered fledging, rather than legacy aerospace giants, reflecting the initiative’s experimental nature.

The first attempt, led by Pittsburgh-based Astrobotic, failed to reach the Moon in January after developing an in-flight fuel leak shortly after separating from its rocket.

It could get another shot later this year, to carry an important water-seeking NASA rover to the south pole, but the green light will depend on the outcome of a detailed review of its first mission.

Intuitive Machines, founded in 2013, is also targeting a region near the south pole. It has two more missions set for this year.

Another Texas company, Firefly Aerospace, has two missions, including one in 2024, with its Blue Ghost lander.

Finally, Massachusetts-based Draper, which built the computer that ran the Apollo spaceships, will attempt to land on the far side of the Moon in 2025.

NASA paid both Astrobotic and Intuitive Machines around $100 million each for their respective first missions.

– Lasting presence –

CLPS deliveries serve the bigger goal of better understanding the environmental risks — from radiation to lunar dust — that face American astronauts when they touch down on the surface no sooner than 2026.

Unlike during Apollo, when the United States was driven by the Cold War to act fast and chose near the equator for five short  trips, NASA is now taking its time to build a “sustained presence,” involving habitats, on the south pole.

Here, it hopes to harvest ice for drinking water and rocket fuel, with an eye on using our cosmic neighbor as a waypoint for crewed missions to Mars.

Share this:
Continue Reading

News desk

In a doughnut in Japan, unlocking the power of the Sun

Published

on

By

To the untrained eye, the JT-60SA looks like a contraption from 1970s sci-fi
Share this:

With its tangle of pipes and pumps leading to a metal pot the size of a five-storey building, Japan’s JT-60SA machine looks to the untrained eye like a contraption from 1970s sci-fi.

But inside it is a doughnut-shaped vessel where experiments done at millions of degrees could help unlock a carbon-free, inexhaustible and safe power source for the future: nuclear fusion.

“Fusion energy, the power behind the Sun and the stars, has been a great prize for energy research for decades, ever since it was first attempted in the 1950s and 60s to find some way to reproduce this power of the Sun here on Earth,” project leader Sam Davis told AFP on a recent tour.

“Not only is (fusion) free from greenhouse gases and free from long-lived nuclear waste, but it’s compact, doesn’t cover the whole landscape, and can generate industrially useful quantities of power,” the British-German engineer said.

Unlike fission, the technique currently used in nuclear power plants, fusion involves combining two atomic nuclei instead of splitting one, generating vast amounts of energy.

The process is safe and there are no nasty by-products like fissile material for a nuclear weapon or hazardous radioactive waste that takes thousands of years to degrade, its proponents say.

– Swirling plasma –

Taking 15 years to build in Naka, northeast of Tokyo, the JT-60SA is 15.5 metres (51 feet) tall and 13.7 metres (45 feet) wide, comprising a so-called tokamak vessel able to contain swirling plasma heated to millions of degrees.

Inside the facility, which was inaugurated in December, the aim is to get nuclei of hydrogen isotopes to fuse into an atom of helium, releasing energy, and mimicking the process that takes place inside the Sun and stars.

“With only one gram (0.04 ounces) of a mixed fuel… we can obtain an energy equivalent to eight tonnes of oil,” said Takahiro Suzuki, deputy project manager for the Japan side of the joint project with the European Union.

But despite decades of efforts, the technology remains in its infancy and is very expensive.

Currently the largest such facility in operation, the JT-60SA is the little brother and guinea pig of the International Thermonuclear Experimental Reactor (ITER) being built in France.

According to media reports, ITER — a project run by six countries and the European Union — is years behind schedule and could end up costing as much as 40 billion euros ($42.3 billion), far more than first projected.

The holy grail of both projects, as well as others around the world, is to develop technology that releases more energy than is needed to fuel it — and at a large scale and for a sustained period.

The feat of “net energy gain” was managed in December 2022 at the National Ignition Facility at Lawrence Livermore National Laboratory in the United States, home to the world’s largest laser.

– ‘Flash in a can’ –

But the US facility uses a different method from ITER and the JT-60SA known as inertial confinement fusion, in which high-energy lasers are directed simultaneously into a thimble-sized cylinder containing hydrogen.

“Magnetic confinement, and in particular, tokamaks, of the kind that JT-60SA is, are much more applicable to running a steady state power plant, to steady energy production as we would need,” Davis said.

“This is not just a flash in a can.”

But with the world record set by China for heating plasma to the required temperature — 120 million degrees Celsius (216 million degrees Fahrenheit) — currently just 101 seconds, there is still a long path ahead.

“Nuclear fusion can certainly contribute to a future energy mix. Exactly on what timescale is very hard to say. It will come down ultimately to how much is invested in the field (and) how much society wants to pursue this as a solution,” Davis said.

Share this:
Continue Reading

Featured