Times Top News

Times Top News

OpenAI mulls watermarking ChatGPT generated text, but treads with caution

OpenAI mulls watermarking ChatGPT generated text, but treads with caution

by | Aug 5, 2024 | Technology


2 min read Last Updated : Aug 05 2024 | 11:06 AM IST

OpenAI has had a system ready for years to watermark text generated by ChatGPT, but it has failed to reach a consensus internally about whether to release it or not – according to the Wall Street Journal.


OpenAI has confirmed that it is working on a text watermarking method, after the report by Wall Street Journal. According to the US-based artificial intelligence company, its text watermarking method is accurate and “even effective against localized tampering, such as paraphrasing, it is less robust against globalized tampering.” However, the company has kept it on hold to date over concerns about stigmatisation of use of AI as a useful writing tool for non-native English speakers.


What is OpenAI’s watermarking method


OpenAI’s watermarking can be described as the process of regulating the model’s predictions of the word and phrase that are going to come up next to create a pattern that can be noticed. Reportedly, the company feels that watermarking is the right thing to do, but also believes watermarking could prevent people from using ChatGPT.


Watermarking can prove to be an effective way to find out if a content was written by AI, and according to reports the company has said that the watermarking does not affect the quality of content produced by ChatGPT.


OpenAI said that it is exploring embedding metadata, too. “For example, unlike watermarking, metadata is cryptographically signed, which means that there are no false positives. We expect this will be increasingly important as the volume of generated text increases. While text watermarking has a low false positive rate, applying it to large volumes of text would lead to a large number of total false positives,” said OpenAI in a blogpost it updated on August 4.


OpenAI had shut down its previous AI text detector, AI classifier over low rate of accuracy. This new tool is expected to be different from the previous ones. Watermarking would focus on detection of content from ChatGPT only, resulting in an invisible watermark within the written content that is detectable by different tools.

First Published: Aug 05 2024 | 11:06 AM IST



Source link

iPhone 16 series may launch without Apple Intelligence in September: Report

iPhone 16 series may launch without Apple Intelligence in September: Report

by | Aug 5, 2024 | Technology


Representative Image: iPhone 15 series (Photo: Bloomberg)

2 min read Last Updated : Aug 05 2024 | 10:52 AM IST


Apple is not pushing the release of the 2024 iPhone series despite delay in Apple Intelligence roll out. According to a Bloomberg report, the iPhone 16-series will be launching in September, in-line with the launch schedule that Apple has followed for the last few years. This comes after Bloomberg’s last week reports that stated the iPhone 16 series devices would not come with Apple Intelligence tools out-of-the-box.


In 2011, Apple postponed its hardware launch event to include iCloud and virtual assistant Siri, which were the new software features then. Apple was expected to do the same this year as the company’s new artificial intelligence features are reportedly not ready yet. However, as per Bloomberg’s new report, Apple is not holding back the release of the new hardware and iPhone 16 series is expected to be unveiled around the same time as last year.


According to the report, new AI features that Apple collectively calls Apple Intelligence would likely roll out in October as part of the iOS 18.1 update. The report also stated that Apple will only be offering limited AI features and tools with the initial update, with the company planning a gradual release schedule for Apple Intelligence. This is also one of the reasons for Apple to not push back the release of the next iPhone models as the headline features such as OpenAI’s ChatGPT integration and an updated Siri with more control are expected to be available on the consumer front next year.


Last week, Apple released the iOS 18.1 developer beta, offering the initial set of Apple Intelligence features for testing to select users. AI features available on the attest developer beta includes Smart Reply in Messages, Writing Tools for summarisation and more, email summarisation and sorting in Mail, audio transcriptions, Memory Movie in Photos and more. The update also includes an updated interface for Siri and improvements across the board.


Though the iOS 18.1 developer beta is available for testing on all eligible iPhone models, the Apple Intelligence features are exclusive to the iPhone 15 Pro and iPhone 15 Pro models.

First Published: Aug 05 2024 | 10:52 AM IST



Source link

Meta in talks to use voices of Judi Dench, Awkwafina, and others for AI

Meta in talks to use voices of Judi Dench, Awkwafina, and others for AI

by | Aug 4, 2024 | Technology


3 min read Last Updated : Aug 04 2024 | 11:27 PM IST


By Mike Isaac & Nicole Sperling


Meta is in discussions with Judi Dench, Awkwafina (pictured), and other actors and influencers for the right to incorporate their voices into a digital assistant product called MetaAI, according to three people with knowledge of the talks, as the company pushes to build more products that feature artificial intelligence.

 


Apart from Dench and Awkwafina, Meta is in talks with comedian Keegan-Michael Key and other celebrities, said the people, who spoke on the condition of anonymity. They added that all of Hollywood’s top talent agencies were involved in negotiations with the tech giant.


The talks remain fluid, and it is unclear which actors and influencers, if any, may sign on to the project, the people said. If the parties come to an agreement, Meta could pay millions of dollars in fees to the actors.

 


A Meta spokesman declined to comment. The discussions were reported earlier by Bloomberg.

 


Meta, which owns Facebook, Instagram and WhatsApp, has invested heavily in artificial intelligence, which the biggest tech companies are racing to develop and lead. Meta has plowed billions into weaving the technology into its social networking apps and advertising business, including by creating artificially intelligent characters that could chat through text across its messaging apps.

 


On Wednesday, Mark Zuckerberg, Meta’s chief executive, increased how much his company would spend on AI and other expenses this year to at least $37 billion, up from $30 billion at the beginning of 2024. Zuckerberg said he would rather build too fast “rather than too late” to prevent his competitors from gaining an edge in the AI race.

 

One area of AI that is rapidly emerging are chatbots with voice abilities, which act as virtual assistants. In May, OpenAI, a leading AI company, unveiled a version of ChatGPT that could receive and respond to voice commands, images, and videos. It was part of a wider effort to combine conversational chatbots with voice assistants like the Google Assistant and Apple’s Siri.

 


OpenAI later suspended the release of its voice-related ChatGPT after the actress Scarlett Johansson, who had provided the voice for an AI system in Spike Jonze’s 2013 movie, “Her,” accused the company of using a voice “eerily similar to mine” despite her refusals to participate in the product. Meta is angling to strike deals with celebrities in a way that avoids ticking off top talent. Under the terms of the proposed contract, Meta would record the voices of these celebrities for potential use in MetaAI, which users could interact with and ask questions across Facebook, Instagram, WhatsApp, and Messenger, as well as Meta’s Ray-Ban augmented reality glasses, the people said. Any deal would be for a set period and could be renewed or termin­ated when the contract was up. Actors would not release the rights to their voices indefinitely.

 


Meta is trying to lock down the deals in time for its Connect techn­ology conference in late September, when the company plans to debut new AI-focused products. 

 


©2024 The New York Times News Service

First Published: Aug 04 2024 | 11:26 PM IST



Source link

Algo Rhythm: Space technology startups soar high, says Tracxn report

Algo Rhythm: Space technology startups soar high, says Tracxn report

by | Aug 4, 2024 | Technology


1 min read Last Updated : Aug 04 2024 | 9:04 PM IST

India has more than 100 space technology startups and most were founded in the past five years. It owns 55 active space assets that include communication, meteorological and Earth observation satellites. According to a report by Tracxn, a market intelligence platform, 2023 was the highest-funded year for such firms. They overall got $126 million, a 7 per cent rise from the $118 million raised in 2022 and an increase of 235 per cent compared to $37.6 million in 2021. In 2024, space tech startups have raised $10.8 million yet.

First Published: Aug 04 2024 | 9:04 PM IST



Source link

Big Tech's AI promises become a 'show me' story for investors: Report

Big Tech's AI promises become a 'show me' story for investors: Report

by | Aug 3, 2024 | Technology


There were some bright spots in the week that signaled the AI trade isn’t completely dead | Representational Image

4 min read Last Updated : Aug 03 2024 | 7:28 PM IST


After a jam-packed week of earnings reports from megacap technology companies one thing is clear: as profits slow, investors aren’t impressed by artificial-intelligence promises anymore. They want to see results.


With six companies inside a group known as the Magnificent Seven already having reported, year-over-year earnings growth has slowed to nearly 30 per cent in the second quarter, down from 50 per cent in the prior period. Analysts expect that rate to decelerate further, to about 17 per cent for those companies in the third quarter.


Results from Microsoft Corp., Meta Platforms Inc., Amazon.com Inc. and Apple Inc. this week signaled that the biggest companies in the world are still heavily investing in artificial intelligence. However, shares of Microsoft and Amazon slid after their reports because of fears that those AI investments aren’t paying off for them — at least not yet — echoing the slip in Alphabet Inc.’s stock a week earlier.

“Investors are entering a ‘show me’ phase, seeking concrete evidence of AI’s impact on revenue and productivity,” said Adam Sarhan, founder and chief executive officer at 50 Park Investments. “This is causing some skepticism and volatility.”


Tesla Inc.’s July 24 report also disappointed investors, while Nvidia Corp. is due to release results later this month. The latest prints and commentary this week added to existing volatility.


Investors had already been shifting from large, trusted stocks into smaller, riskier parts of the market to lessen exposure to Big Tech. The earnings results, combined with the Federal Reserve signaling that a September rate cut may be on the table and a weaker-than-expected jobs report sent the Nasdaq 100 Index spiraling.


On Friday, the tech-heavy index closed down 11 per cent from its July peak, entering a correction. Investors fled AI stocks and bid up bonds, sending Treasury yields lower.


The bond market is “telling us we’re going to have to bring this sucker down real fast, and that’s kind of worrying everybody,” said Kim Forrest, chief investment officer at Bokeh Capital Partners. “Lower interest rates work for equities, except when it’s being done in a hurry because things are bad.”

Amazon’s results, alongside reports from consumer names like McDonald’s Corp. and Starbucks Corp. signaled a weakening US consumer, adding to concerns about a weaker macroeconomic backdrop, she said.


Investors were already concerned about hype-versus-reality in the tech sector, which contributed to sharp reactions when major companies underperformed, said Burns McKinney, managing director and senior portfolio manager at NFJ Investment Group.


“Some of the earnings results that have come in over the last couple of weeks have reminded investors that there’s a lot of really high expectations baked into these valuations,” he added.


There were some bright spots in the week that signaled the AI trade isn’t completely dead.


Investors cheered Meta’s results, including comments from Chief Executive Officer Mark Zuckerberg that signaled investments in AI helped drive targeted ad sales. Advanced Micro Devices Inc. spurred a Wednesday rally in chip stocks after it gave a rosy revenue forecast.


“Essentially what companies are saying is that they have to do this and if they don’t, they could risk being irrelevant in the future,” Gene Munster, managing partner of Deepwater Asset Management, said of the increased capital expenditures on artificial intelligence.


The sharp market reaction doesn’t necessarily mean the AI trade is over, Sarhan said.


“Instead, it suggests a recalibration of expectations,” he said. “We’re seeing a shift from pure hype to a demand for tangible results.”


(Only the headline and picture of this report may have been reworked by the Business Standard staff; the rest of the content is auto-generated from a syndicated feed.)

First Published: Aug 03 2024 | 7:28 PM IST



Source link

ShareChat expands its convertible debentures round to million

ShareChat expands its convertible debentures round to $65 million

by | Aug 3, 2024 | Technology


2 min read Last Updated : Aug 03 2024 | 5:50 PM IST


ShareChat (Mohalla Tech Pvt Ltd), India’s largest homegrown social media company, said it has expanded its convertible debentures round to $65 million, to accommodate new investment from EDBI, a leading Singapore-based global investor. 


These additional funds will be utilised to invest further in the company’s advertising tech stack and expand the consumer transactions businesses with investment in newer monetisation features for the creator ecosystem. 


Currently, the ShareChat app is operationally profitable for the past several months, while the Moj app is close to operating break-even.


In April this year, ShareChat raised $49 million via convertible debentures in the funding round led by existing investors Lightspeed, Temasek, Alkeon Capital, Moore Strategic Ventures and HarbourVest, amongst others.


According to media reports, Sharechat is closing a $16 million fundraise, while laying off about 5 per cent of its workforce (around 30-40 employees) after a bi-annual performance review.


“The fact that ShareChat could attract new investment from a sovereign fund like EDBI is a significant positive sign,” said a person familiar with the matter.  


“Attracting new investors of this calibre in today’s cautious private investment environment reflects that the market is taking note of the company’s progress towards building a profitable business,” the person said.


ShareChat  runs India’s largest short video platform, Moj in addition to the ShareChat app, which collectively cater to over 325 million users. ShareChat said it has democratised social media in India by offering content in over 15 Indian languages. 


Through its multilingual approach and diverse ad formats, the company has worked for more than 200 advertisers in the country including HUL, Maruti, P&G, Hero and Samsung. 


Over the past couple of years, ShareChat said it has rapidly grown its monetising avenues, including virtual gifting where it stands as the market leader by a huge margin.

First Published: Aug 03 2024 | 5:50 PM IST



Source link

About-us

Privacy Policy

Email: hansrajagrawalakola@gmail.com

YouTube
Instagram
WhatsApp