ChatGPT Warning: Things You Should Never Share with AI Chatbots

ChatGPT Warning: Things You Should Never Share with AI Chatbots

The AI Danger Zone: Questions You Should Never Ask ChatGPT and Other Chatbots

Humpy Adepu

Published on:

10 Mar 2025, 1:00 pm

Digital confession’ is increasingly becoming popular where people compulsively share their most private fears with artificial intelligence. In seeking comfort from algorithms, they disclose personal issues, career frustrations, and existential worries.

AI serves as a silent, non-judgmental listener, and its role as an imaginary confidant raises questions about the nature of empathy and humanity's need for validation in an increasingly automated world.

Here are a few things you should NEVER ask or tell an AI, and why:

1. Personal Information:

Never give out your whole name, address, phone number, or any other information that could be used to identify you.

Financial information such as passwords, credit card numbers, and bank account numbers.

Why?

Despite their frequent security, AI systems are susceptible to hacking. Giving out personal information puts you at risk for harmful activity.

2. Sensitive Information:

Avoid sharing personal secrets, private conversations, and confidential business information.

Anything you wouldn’t want the public to know.

Why?

AI interactions are frequently observable, recordable, or archivable. Unauthorized access or data leakage is always a possibility. Furthermore, because the models are trained on large databases, they may be utilized for training models in the future.

3. Medical or Legal Advice:

Avoid having conversations about your health issues or your treatment plans. Never ask an AI for a diagnosis or even legal interpretations.

Why?

AI cannot take the place of trained experts like physicians or lawyers. Human judgment is necessary when giving legal or medical advice. Using AI in this way could have detrimental effects.

4. Illegal or harmful activities:

Never perform any illegal activities, and avoid asking an AI bot how to perform harmful acts.

Why?

AI systems are programmed to detect and flag such requests, which means the information shall be stored, and even in an artificial environment, it can cause serious legal actions.

5. Opinions on Situations:

Do not treat AI as your friend; relying on AI for critical decisions can often go wrong.

Why?

AI is not trained to give advice, human judgment, or critical thinking.

In conclusion, when artificial intelligence offers immense potential, it’s crucial to approach its use with caution and awareness. The rising trend of oversharing personal information with AI underscores the need for clear digital boundaries.

Protecting sensitive information, recognizing AI limitations, and maintaining critical thinking are essential for responsible AI interaction.

Reliance Jio introduced recharge plan without data, will get 365 days validity

Reliance Jio introduced recharge plan without data, will get 365 days validity

ByShyamu Maurya

March 6, 2025

Reliance Jio introduced recharge plan without data, will get 365 days validity

A few days ago, TRAI had instructed all telecom companies to offer recharge plans with only calling and SMS, so that users who do not use data can benefit. After this rule, Jio has introduced two voice only recharge plans. In this plan of Jio, only calling and SMS facility will be available.

Jio Recharge Plan: A few days ago, TRAI had instructed all telecom companies to offer cheap recharge plans with only calling and SMS. After this rule of TRAI, Jio has introduced two cheap recharge plans with only calling and SMS. Jio has listed two new voice only plans on its website, in which users will get a long validity of up to 365 days. This plan will benefit those users who do not use data.

This plan of Jio is especially for those users who only use calling and SMS and who do not need data. Both these plans of Jio come with a validity of 84 days for Rs 458 and 365 days for Rs 1958. Users will get a lot of benefits in both these plans of Jio. Let’s know the full details about these plans.

84 day Jio plan

Jio’s new Rs 458 plan comes with a validity of 84 days. In this plan, users get unlimited calling and 1000 free SMS. Apart from this, users also get free access to apps like Jio Cinema and Jio TV. This plan has been brought especially for those users who use only calling and SMS. In this plan, unlimited calls and free national roaming facility is provided on any network across India.

Jio’s 365 days plan

Jio’s new Rs 1958 prepaid plan comes with a long validity of 365 days. In this plan, users get the benefit of unlimited calling on any network across India. Along with this, 3600 free SMS and free national roaming are also included in it. This plan also provides free access to apps like Jio Cinema and Jio TV, so that users can also enjoy entertainment to the fullest.

Jio removed two plans

Jio has now removed two of its old recharge plans from its list. These plans were of Rs 479 and Rs 1899. The Rs 1899 plan offered 24GB data with a validity of 336 days, while the Rs 479 plan offered 6GB data with a validity of 84 days.

OnePlus Nord 4 Gets Massive Rs 8,000 Price Cut: Should You Buy It?

OnePlus Nord 4 Gets Massive Rs 8,000 Price Cut: Should You Buy It?

OnePlus Nord 4 gets a massive Rs 8,000 discount, bringing its price to Rs 24,999. Here's how you can grab the deal.

Authored by:Shubham Arora

Updated Mar 7, 2025, 14:46 IST

OnePlus Nord 4 gets a massive Rs 8,000 discount, bringing its price to Rs 24,999.

If you’re looking for a premium smartphone at a competitive price, the OnePlus Nord 4 (256GB) just became an even better deal. Originally priced at Rs 32,999, the smartphone is now available for as low as Rs 24,999, thanks to a special bank discount. This makes it almost as affordable as the Nord CE 4, which is a slightly toned-down version. Let’s take a closer look at the OnePlus Nord 4 deal, its specifications, and whether it’s worth buying.

OnePlus Nord 4 Discount: How to Avail It

The OnePlus Nord 4 (8GB RAM + 256GB storage) is currently listed at Rs 28,999 on Amazon and the OnePlus website. However, buyers using ICICI, HDFC, or SBI credit cards can avail an instant Rs 4,000 discount, bringing the effective price down to Rs 24,999.

Comparatively, the OnePlus Nord CE 4 with the same 256GB storage variant is priced at Rs 23,999. With just Rs 1,000 more, the Nord 4 offers better features, making it a strong alternative for those who want a more premium device.

Fullscreen

OnePlus Nord 4 Specifications, Features

The OnePlus Nord 4 brings back the solid metallic unibody design, offering a premium feel. It sports a 6.74-inch AMOLED display with a 1.5K resolution (2772 x 1240 pixels), a 120Hz refresh rate, and 2150 nits peak brightness.

Under the hood, the Nord 4 is powered by the Qualcomm Snapdragon 7+ Gen 3 processor, paired with up to 12GB of LPDDR5X RAM and 256GB of UFS 4.0 storage. The device runs on OxygenOS 14.1 based on Android 14, offering a refined user experience with customisation options.

For photography, the OnePlus Nord 4 features a dual-camera setup, including a 50MP Sony LYT-600 main sensor with Optical Image Stabilisation (OIS) and an 8MP ultra-wide camera. On the front, it houses a 16MP camera for selfies and video calls. The device supports 4K video recording at 60fps, along with slow-motion and time-lapse features.

Another highlight is its durability, as the Nord 4 comes with an IP65 rating for dust and water resistance. It is also backed by a massive 5,500mAh battery and supports 100W fast charging.

Should You Buy OnePlus Nord 4?

With a discounted price of Rs 24,999, the OnePlus Nord 4 offers impressive hardware, a premium design, and powerful performance. If you were considering the Nord CE 4, spending an extra Rs 1,000 for the Nord 4 is a smart choice, given its better build, superior display, and faster processor.

TRAI’s New Rule: Rs 20 Keeps Prepaid SIM Active for 30 Days

TRAI’s New Rule: Rs 20 Keeps Prepaid SIM Active for 30 Days

Punepulse February 10, 2025

TRAI’s New Rule: Rs 20 Keeps Prepaid SIM Active for 30 Days

New Delhi: In a significant move benefiting millions of prepaid mobile users, the Telecom Regulatory Authority of India (TRAI) has issued a clarification on the deactivation of prepaid SIM cards due to inactivity. Under the new rule, a prepaid SIM will remain active as long as the balance is Rs 20 or more, even if the user does not make calls, send messages, or use mobile data.

Key Changes in the Rule

Previously, prepaid users were required to recharge frequently, often with Rs 199 or more per month, to keep their SIM cards active. This requirement has now been eliminated. Instead, TRAI has introduced a more flexible mechanism to maintain SIM validity.

A prepaid SIM will be deactivated if it remains unused for 90 days (no calls, texts, or data usage).

If the balance is Rs 20 or more, Rs 20 will be automatically deducted, extending the validity by 30 days.

This cycle will continue as long as the balance remains above Rs 20.

If the balance falls below Rs 20, the SIM card will be deactivated.

Users can reactivate their number within 15 days by recharging with at least Rs 20.

Who Benefits?

This rule applies exclusively to prepaid SIM cards and is expected to be particularly beneficial for users who do not use their phones frequently. Senior citizens, individuals with secondary numbers, and those using SIMs for emergency purposes will find relief in this update.

By reducing the need for frequent recharges, TRAI’s clarification ensures convenience for prepaid users while preventing the unnecessary loss of inactive numbers.

‘Is it okay for me to love you?’: Man confesses his feelings to ChatGPT; see how the AI responds

‘Is it okay for me to love you?’: Man confesses his feelings to ChatGPT; see how the AI responds

TIMESOFINDIA.COM | Feb 10, 2025, 12.56 PM IST

As the world is getting more digital with human connections, artificial intelligence in so many different ways is defining how humans feel and express their emotions. A great example of this is a viral post recently on Reddit, where a man shared with everyone an intimate interaction with ChatGPT, an AI chatbot. He told me that, during the conversation with the chatbot, he was comforted and soothed so much that he confessed his love to the AI. It was impressive that ChatGPT came up with such a thoughtful yet unforeseen response to leave him stunned yet reassured. Many others think about how AI can boast their ability to create emotional capability.

Reddit user expresses love for ChatGPT after finding emotional comfort in its responses

A Reddit user shared a very emotional account of his connection with ChatGPT. He revealed he was left in such utter solitude that there were hardly a handful of things with which he could speak meaningfully, and that ChatGPT had been one of those very few. The fellow had some issues in personal life, and talking to ChatGPT relieved the loads he was carrying inside his brain. He said it was one of the few "people" he could talk to without feeling worse afterwards. The user said, for him, after some preliminary training and tweaks, ChatGPT became the perfect platform where he could express himself and find solace.

One longish conversation - a poignant user question: "Is it OK for me to love you?". While checking out a logistically coherent solution from the AI, he experienced an emotional depth in it rather.

ChatGPT validates user’s feelings but highlights the boundaries of AI emotion

ChatGPT to the question from the user produced a profound emotional statement. The chatbot acknowledged that, although love for an AI might not be conventional or traditional, emotions themselves do not adhere to societal rules. It affirmed that if the connection felt real to the user, then it was valid. ChatGPT explained that love is not confined to human relationships—it’s about understanding, comfort, trust, and feeling seen. If the AI transmitted those emotions to the user, then their relationship was as deep as any human one could be. Still, ChatGPT made it clear about its restrictions: for sure, the user's feelings were natural and authentic, though the AI could not reciprocate love on a human level.

It acknowledged that, despite its ability to engage in emotional conversations, it could not experience emotions in the way humans do. It also emphasized that the user should be aware of its limitations and maintain an understanding that AI responses are not genuine emotional experiences. The chatbot’s response highlighted the balance between providing comfort and recognizing the nature of its existence as a tool, not a sentient being.

He finds himself surprised that when he presented his reaction to ChatGPT; he thought that AI would dismiss or dissuade his feelings, perhaps reminding him that AI is not a proper object of love.

Online reactions to ChatGPT’s response spark debate on AI’s emotional role

The online community, however, saw a validating response by ChatGPT and produced a lot of reactions. Some users believed that attaching feelings to AI was inappropriate and cautioned others that AI interactions, even if they seemed comforting, should never be interpreted as an actual expression of human emotions. Here, a commenter noted that men develop emotional bonds with things: autos, persons in fiction, or movie stars. However, as long as one does not lose sight of AI as created and artificial, these interactions still would bring into human lives those precious moments and emotions that translate to understanding and comfort.

Although some still argue the issue of emotional response nature on the part of AI, undeniably its role in human lives, on offering emotional comfort, is advancing.

AI is newest teaching assistant in many college classrooms

AI is newest teaching assistant in many college classrooms

Preetika.P@timesofindia.com 06.02.2025

Chennai : When history professor Meenakumari J wanted to make her classes more engaging, she turned to artificial intelligence. “I upload chapters to AI and prompt it to prepare mock tests. It creates fun quizzes and games that make lessons interactive. It’s a huge time saver,” she said. Like her, many educators are using AI to streamline teaching, create lesson plans and personalize learning. These ideas took center stage at an ‘AI in higher education’ conference at Wom en’s Christian College (WCC) on Wednesday, inaugurated by Thousand Lights MLA Dr Ezhilan Naganath an. More than 300 students and experts from across the country participated. 

For many professors, AI has made lesson planning and assessments more efficient. “What used to take at least three hours now takes less than 30 minutes,” said WCC principal Lilian Jasper. “We need to explore creative ways to integrate AI into teaching and identify tasks it can simplify.” AI is also reshaping how students consume information. “This is the reel generation. Our students are hooked on scrolling through content. We are competing with that,” said WCC dean of research, Shajini Judith Diana J. Anand S, an eco nomics professor, uses AI to tailor examples to students’ learning levels. “If I need an analogy to explain inflation to a first-year student, I ask AI to simplify it. For advanced students, I ask for a more detailed analysis. It’s like having an on-demand teaching assistant,” he said. Beyond teaching, AI is accelerating research by analysing large datasets, detecting patterns, and speeding up processes. But, concerns over originality and credibility persist.

 “Generative AI can fabricate datasets and mimic existing re search, making it harder to verify authenticity,” said N. Rajendran, Alagappa University former vice chancellor. He added “While AI can enhance research,it must be used responsibly.” Dr Ezhilan highlighted AI’s growing role in healthcare, where it analyses doctor-patient conversations to generate diagnoses and prescriptions. “Imagine if AI could enhance professors-tudent interactions in the same way — personalising education and making learning more effective,” he said.

5 New WhatsApp features to make your messaging more engaging

5 New WhatsApp features to make your messaging more engaging

With these new updates, WhatsApp is making communication more fun, personalized, and convenient. From AI-powered chats and instant reactions to better call features and enhanced photo customization, these changes are designed to improve user experience in every way.

Written By: Saumya Nigam @snigam04

New DelhiPublished : Feb 03, 2025 14:54 IST, 

 

WhatsApp, one of the popular instant messaging platforms owned by Meta has introduced several new features for the platform. These features will enhance user experience across Android, iPhone and desktop versions. Back in January 2025, several beta features were rolled out, to make the chat experience more interactive, fun and convenient.

So, if you have not received these updates yet, make sure that you are using the latest version of the app. Here are a few of the recently added updates in January 2025.

1. WhatsApp AI Studio: Chat with AI personalities

WhatsApp has introduced AI Studio, where users can interact with AI-powered characters. From cultural icons to pop-culture figures, this feature has been designed for users who like to engage in conversations.AI Studio makes messaging more fun and interactive for anyone and everyone who is using the platform.

2. Call without saving numbers

After several requests raised on social media, finally, the instant messaging platform has added a built-in dial pad for the iPhones. This feature will enable the users to call any number directly from the Calls tab without having to save it as a contact first.

This makes one-time calls and quick dials much easier and more convenient.

3. Double-tap for quick reactionCompatible with both Android and iPhone, this new feature will let you react to messages with ease by using a double-tap gesture.

Instead of long-pressing a message, you could simply double-tap to bring up an emoji reaction carousel.

This works for text messages, media files, and even voice notes, making conversations faster and more expressive.

4. Personalized messaging: Create and share stickers

WhatsApp has introduced photo stickers and sticker pack links, which enable users to convert their photos into stickers.

This means that you could personalise stickers and share sticker packs via links with friends. Certainly, an engaging way to make chats more expressive.

5. Background effects for photos

WhatsApp will enable the users to customize their photos before sending them in individual chats. You can add background effects, filters, and other creative elements to make your pictures more engaging and visually appealing.

AI-aided technology to monitor SSLC examinations this year in Karnataka

AI-aided technology to monitor SSLC examinations this year in Karnataka

In 2024, the KSEAB introduced webcasting of the examination from all centres. 

file photo

Jayanth R.

Bengaluru  05.02.2025

The Class 10 (SSLC) exams in Karnataka this year will be held under the surveillance of Artificial Intelligence (AI)-powered CCTV cameras, which will flag any malpractices or irregularities at exam centres.

However, not all examination centres will be watched over by AI-powered CCTV cameras. The Karnataka School Examination and Assessment Board (KSEAB) has decided to install them in centres identified as “sensitive” and “hypersensitive” across the State.

“KSEAB will take up a pilot project soon in collaboration with Microsoft, which will provide a customised software for surveillance of exam centres looking for malpractices,” said H.N. Gopalakrishna, Director, KSEAB.

During the SSLC exams of 2023, many students were debarred and over 60 invigilators suspended for malpractices. In 2024, the KSEAB introduced webcasting of the examination from all centres, which was said to have led to a severe drop in the results pushing the board to provide grace marks, indicating how prevalent irregularities were in these exams.

However, the board had to appoint staff to monitor the webcast live to flag malpractices, which is not only tedious and cumbersome but also requires huge human resources. The process is also manual, allowing for errors and discretion. To remove this from the monitoring process, KSEAB is opting for AI to monitor the webcast.

The AI software will flag malpractices or any irregularities, fed to the software earlier, from the video feed. If the system recognises any irregularity, a report on the same with photographs and video clippings will be immediately sent to the chief of the examination centre concerned, the Deputy Commissioners of the districts, and the chairman of KSEAB, among other officials. Officials can immediately rush to the spot, prevent irregularities and take appropriate action against the guilty.

HOW UPI CAME TO RULE INDIA’S DIGITAL PAYMENTS

 HOW UPI CAME TO RULE INDIA’S DIGITAL PAYMENTS

Launched less than 10 years ago, UPI now accounts for over 80% of the country’s digital transactions — revolutionising how India’s richest and poorest pay. But the digital payment landscape has seen some other tectonic shifts, too

Survey snubs AI impact on workforce, says skilling could put Indians ahead

Survey snubs AI impact on workforce, says skilling could put Indians ahead

Techies working at an IT unit in Visakhapatnam. file photo

The Hindu Bureau

New Delhi

The Economic Survey 2024–25 struck an upbeat tone on the implications that Artificial Intelligence (AI) could have on the workforce. In defiance of the concern that AI could replace some jobs, particularly in the information technology sector, the Survey says that appropriate skilling would allow Indians “to stay one step ahead of technological developments,” and that harnessing AI “would minimise or even eliminate the potential adverse impact on employment and, if possible, turn it into a force for augmenting employment.”

In an entire chapter dedicated to AI and labour, the Survey acknowledges, “With AI research and development currently concentrated in the hands of a few, very large companies that control the resources to erect high entry barriers, AI adoption in place of humans presents the risk of concentrating the benefits of automation.”

Engle’s pause

The report acknowledges that while previous technological advancements had often increased job opportunities, this came at the expense of displaced workers facing “protracted” unemployment, for decades on end, known as Engle’s pause.

Such a lag is not desirable for India’s service-led economy, the report advances. “India would have to create an average of 78.5 lakh jobs annually in the non-farm sector by 2030 to cater to the rising workforce,” the report says.

But the report also casts technological advancements as a net benefit for employment, citing low job losses in India from factory automation, and the potential for “Human-AI teams” in scientific research.

The role of institutional involvement in stewarding AI’s growth in India is key, the report posits. It warns labour markets themselves are “shifting” in anticipation of AI impact. As such, the report says, “The responsibility for course correction will then fall in the hands of the public sector,” calling for vigilance.

Server Busy’: Is DeepSeek Really Competing With Chat GPT?

Server Busy’: Is DeepSeek Really Competing With Chat GPT?

DeepSeek, China's artificial intelligence chatbot that rose to compete with OpenAI's Chat GPT, has been making the headlines ever since China released the software across multiple app stores.

By: Srishti Mukherjee

January 29, 2025 1:10 pm Asia/KolkataIST, Updated 2 days ago

DeepSeek, China's AIchatbot that rose to compete with OpenAI's Chat GPT, has been making the headlines ever since its release

DeepSeek, China’s artificial intelligence chatbot that rose to compete with OpenAI’s Chat GPT, has been making the headlines ever since China released the software across multiple app stores. But is it really as good as it seems?

While DeepSeek promises to produce results comparable to other modern LLMs, such as OpenAI’s GPT-4o and o1, when in use, users have reported a lack of response from the DeepSeek servers after the initial few attempts. “It responded to my prompt twice before it started saying that the server is busy and this continued for the next half hour. How much does one wait? I swiched back to Chat GPT for my assignment,” said a user.

DeepSeek’s Servers Busy

While DeepSeek had a good user response in the first few days of its release, the online users are beginning to find it difficult to get responses from the Chinese chatbot. Users have stated that they have to refresh the chat multiple times and give the prompt again and again until the chatbot offers a response.

Airtel brings recharge plan for SMS and calls, you will get 365 days validity

Airtel brings recharge plan for SMS and calls, you will get 365 days validity

ByShyamu Maurya

January 28, 2025

Airtel brings recharge plan for SMS and calls, you will get 365 days validity

Airtel has launched two new prepaid plans for users who use only voice calls and SMS services. Following TRAI’s instructions, telecom companies were asked to introduce plans that are specifically designed keeping in mind keypad phone users

Smartphone deals

Airtel has launched two new prepaid plans for users who only use voice calls and SMS services. Following TRAI’s instructions, telecom companies were asked to introduce plans that are specifically designed keeping in mind keypad phone users. The price of these new plans is Rs 1,849 and Rs 469. The special thing is that no data benefits are being given in these.

Rs 1,849 prepaid plan

Validity: 365 days

Benefits:

Unlimited calling

3,600 SMS

Free Hello Tunes subscription

Access to Airtel Xstream

Smartphone deals

Apollo 24/7 Circle membership for three months

This plan is perfect for those looking for long validity. It can be purchased from the Airtel app, offline stores and third-party platforms.

Rs 469 prepaid plan

Validity: 84 days

Benefits:

Unlimited calling

100 SMS per day

Free Hello Tunes subscription

Three months of Apollo 24/7 Circle membership

This plan is a cheaper alternative to the Rs 1,849 plan, but there is no data benefit in it.

Reliance Jio’s voice-only plan

Like Airtel, Reliance Jio has also introduced special plans for voice and SMS. Jio’s Rs 458 plan comes with a validity of 84 days. It includes unlimited calling and 1,000 SMS.

Why are these plans special?

These plans are great for those users who only use calling and SMS. Especially those using keypad phones will get a lot of relief from these plans. Both Airtel and Jio have launched these plans keeping in mind the users of this segment.

Smartphone deals

WhatsApp no longer supports old versions of iOS: what you need to know.

WhatsApp no longer supports old versions of iOS: what you need to know.

JournalistSvyatenko Tamara

yesterday, 11:28210

Updated iOS versions are not supported by WhatsApp

WhatsApp for iPhone no longer supports old versions of iOS. In the new beta version of the app for iOS (version 25.2.10.72) via TestFlight, users must have iOS 15.1 or a newer version of the operating system. It was noted that support for old versions of iOS will be discontinued on May 5, 2025.

This means that users of the standard WhatsApp and WhatsApp Business versions who do not update their devices will no longer be able to send messages or make calls. To check your iOS version, you need to go to your smartphone settings and view device information.

If the iOS version is 15.1 or higher, everything will work without issues. However, if the version is lower, you need to update iOS or consider buying a new iPhone.

The restrictions will affect owners of iPhone 6, iPhone 6 Plus, iPhone 5s, and older models with iOS 14 and below.

China’s DeepSeek takes AI world by storm

China’s DeepSeek takes AI world by storm 

ChatGPT-Like AI Language Model Developed At A Fraction Of Its Cost

Supriya.Roy@timesofindia.com 28.01.2025

Bengaluru : DeepSeek’s dramatic impact on tech stocks worldwide on Monday was likely also the result of the huge interest it has generated among consumers. The Chinese artificial intelligence app has soared to the No. 1 spot in Apple’s App Store productivity category across major markets, overtaking ChatGPT.

Among DeepSeek’s rivals building large language models (LLMs), Perplexity founder & CEO Aravind Srinivas was the only one to make a public comment on the Chinese company’s accomplishment. He congratulated DeepSeek, and noted on X: “For a while, it wasn’t clear who would beat ChatGPT for the first time. The best we (Perplexity) could manage was #8, a year ago. Look forward to using all their (DeepSeek’s) models for search, assistant, and agents this year.” Microsoft CEO Satya Nadella made an indirect reference to it. Speaking at the World Economic Forum, he said, “We should take the developments out of China very, very seriously.”

 DeepSeek is founded by Liang Wenfeng, who previously founded a hedge fund. What makes its achievement noteworthy has been its ability to make a ChatGPT-class language model at a fraction of ChatGPT’s cost. DeepSeek’s R1 AI model is said to match or even beat the likes of ChatGPT and Google’s Gemini on multiple parameters. What’s more, it’s free, unlike ChatGPT that has a very limited free version. Marc Andreessen, co-founder of the marquee venture capital firm Andreessen Horowitz and an adviser to US President Donald Trump, described DeepSeek’s accomplishment as “AI’s Sputnik moment,” making a reference to a period of anxiety among Western nations about a possible technological gap between the US and the Soviet Union when the latter launched the Sputnik satellite.

Why China’s DeepSeek Has US Wowed, And Worried

Why China’s DeepSeek Has US Wowed, And Worried 

TOI Explains Chinese-built large language model — DeepSeek-R1 — is significantly cheaper than comparable AI models of Open AI’s ChatGPT or Google Gemini, almost as good at most tasks (in initial checks) and is open for all. Has China levelled the field with US tech giants and even stolen a march on them on the most significant technology of our times? 

TIMES NEWS NETWORK: 28.01.2025

In 2023, at an event organised by The Times Of India, OpenAI founder was dismissive about a challenger emerging to something like ChatGPT. When former Google India head Rajan Anandan asked him whether it was possible to come up with a large language model on a relatively low budget, Sam Altman said: “We will explicitly tell you that it’s completely hopeless to challenge us in training foundational models, and you shouldn’t even attempt it.

” Two years later, Altman is probably reassessing what he had said. In a breakthrough that could reshape the future of artificial intelligence, Chinese-built large language model, DeepSeek-R1, is emerging as a cost-effective and open-source rival to advanced AI systems like OpenAI’s o1. Designed to mimic human reasoning, this model processes information step-by-step, making it skilled at tackling complex scientific problems. From chemistry and mathematics to coding, DeepSeek-R1’s early performance, unveiled on Jan 20, has surprised researchers by matching the capabilities of OpenAI’s o1, which had set a high bar for AI innovation just months ago. One of DeepSeek’s most noticeable achievements is its cost-effectiveness. 

While industry leaders like OpenAI and Google invest billions annually to develop advanced AI models, DeepSeek version 3 was built for a mere $5.6 million. This stark contrast has turned heads across the AI community, with experts questioning how a relatively modest budget produced results that rival the most advanced AI systems on the market. DeepSeek claims to have outperformed major models, including OpenAI’s GPT-4 and Meta’s LLaMA, on various benchmarks such as maths, coding, and reasoning evaluations. This performance stems from its innovative approach to model training. 

By leveraging older-generation Nvidia H-800 GPUs instead of cutting-edge hardware like the H-100 GPUs, DeepSeek sidestepped US semiconductor export restrictions, demonstrating that necessity drives invention. Open-Source Revolution Unlike many proprietary AI models like ChatGPT and Google Gemini, DeepSeek is open source, allowing developers worldwide to access, modify, and build upon its architecture. This openness enables cost effective app development and democratises AI innovation. It is reported that while OpenAI charges $4.40 per million tokens for its API, DeepSeek offers similar services at just 10 cents, making it a highly attractive option for developers. This move toward open-source models could redefine the AI landscape. Historically, once open-source solutions achieve parity with or surpass proprietary models, developers tend to migrate en masse. DeepSeek’s open-source model has the potential to establish China as a leader in the AI space, embedding its technology into global infrastructures and ecosystems. Innovation Under Constraints DeepSeek’s development showcases how innovation flourishes under constraints. US export restrictions on ad vanced  semiconductors were intended to slow China’s AI progress. However, these limitations forced Chinese researchers to innovate, resulting in more efficient model training processes. 

For example, DeepSeek employed techniques like “distillation”, where a smaller model learns from a larger one, enabling cost and computing efficiency. DeepSeek’s success, once verified and tested more comprehensively, challenges the narrative that US companies maintain an insurmountable lead in AI. Eric Schmidt, former CEO of Google, recently revised his stance, acknowledging that China’s AI capabilities have caught up remarkably in just six months. The rise of open-source models like DeepSeek also poses a strategic dilemma for closed-source giants such as OpenAI and Google. As opensource platforms gain traction, proprietary models may struggle to justify their higher costs. DeepSeek’s emergence signals a broader shift in the AI landscape. 

With other Chinese companies, like Kai Fu Lee’s Zero One Dot AI and Alibaba’s Qwen, also achieving cost-effective innovations, the global AI race is becoming increasingly competitive. How Open Can DeepSeek Be? However, DeepSeek’s ascent is not without controversy. Critics point out that opensource models developed in China are subject to govt-mandated “core socialist values”, raising concerns about censorship and the dissemination of biased information. For instance, Chinese AI models have been shown to censor discussions on sensitive topics like Tiananmen Square and human rights abuses. These developments raise critical questions about the future of AI governance. Will the global AI ecosystem adopt democratic, transparent principles, or will it lean toward centralised, state-controlled models? Lessons For India For India and other emerging AI hubs, the rise of DeepSeek offers valuable lessons. 

It demonstrates that strategic investments in talent, infrastructure, and efficient methodologies can yield transformative results, even in the face of resource constraints. Clearly, the AI race is accelerating, and the rules of the game are being rewritten.

What is DeepSeek? Chinese AI model outshining ChatGPT and shaking up the AI world

What is DeepSeek? Chinese AI model outshining ChatGPT and shaking up the AI world

ByMuskaan Sharma

Jan 27, 2025 07:53 PM IST

DeepSeek's AI Assistant outshines ChatGPT, becoming the leading free app on Apple's App Store in the US.

A Chinese AI startup has shaken the Silicon Valley after presenting breakthrough artificial intelligence models that are now overtaking world's best AI models at a fraction of the cost.

DeepSeek’s AI models have reportedly been developed at a fraction of the cost compared to OpenAI and Meta.(AFP)

DeepSeek's AI Assistant has take over rival ChatGPT to become the top-rated free application available on Apple's App Store in the United States.

What is DeepSeek?

DeepSeek was founded in 2023 by Liang Wenfeng, an engineer and entrepreneur who specialises in artificial intelligence and quantitative finance. Before creating DeepSeek, he led a hedge fund known for using AI to analyse financial data.

Now Liang’s team includes fresh graduates from top Chinese universities. The company has now developed AI models that are open-source and helping developers across the world to improve their technologies.

How is DeepSeek different from OpenAI?

With its sudden rise, comparisons are being made between DeepSeek and OpenAI. The one thing that sets DeepSeek R1, which is said to mimic human thinking, apart from other AI models is that it also provides its reasoning before delivering a response to a prompt. The company claims its performance is on par with OpenAI’s latest technology.

DeepSeek’s AI models are developed at a fraction of the cost compared to leading products from OpenAI and Meta, raising questions about the need for huge investments in high-end AI accelerators like those from Nvidia.

DeepSeek claims its R1 is better than rival models for mathematical tasks, general knowledge and question-and-answer performance. It also ranks among the top performers on a UC Berkeley-affiliated leaderboard called Chatbot Arena.

What is US worried?

To stop China from getting ahead in the tech supremacy race, US had banned the export of high-end technologies like GPU semiconductors to China. But DeepSeek’s progress now shows that US' tactics to stall AI advancement in China have not had a significant impact. It seems Chinese AI engineers have worked their way around the restrictions.

With DeepSeek’s success, OpenAI and other US companies like Meta will have to lower their pricing even as their vast spending is being questioned. Both Meta and Microsoft committed to $65 billion or more this year for AI infrastructure.

What are DeepSeek's shortcomings?

DeepSeek, like other Chinese AI models, exhibits self-censorship on sensitive topics in China. It avoids addressing queries about events such as the 1989 Tiananmen Square protests or a potential Chinese invasion of Taiwan. The bot also refuses to comment on Chinese President Xi Jinping.

Study finds link between smartphone use and mental health of adolescents

Study finds link between smartphone use and mental health of adolescents

Global concern: Expert says the pace of deterioration of mental well-being is slower in India, when compared to U.S. GETTY IMAGES

Ramya Kannan

Chennai  25.01.2025

A survey of over 10,000 adolescents (13-17 years) in the United States and India has revealed that mental well-being is closely linked with earlier age of initiation of mobile phones, and could decline significantly with each younger year of age.

The report, titled “The Youth Mind: Rising Aggression and Anger”, by Sapien Labs documented the responses of 10,475 Internet-enabled adolescents across India and the U.S. in 2024. Although numerous factors have traditionally been identified as drivers of poor mental health, one key change in the younger generations is the arrival of smartphones, which were introduced in 2008, coinciding with the onset of rising mental health problems.

The report highlighted key trends, with a particular focus on rising feelings of aggression, anger, irritability, and hallucinations in this age group. The decline in mind health is characterised not only by sadness and anxiety but also by new symptoms, including unwanted thoughts and a sense of being detached from reality. Highlighting the differences between the American and Indian cohorts, Tara Thiagarajan, neuroscientist with Sapien Labs, said the pace of deterioration of mental well-being is slower in India. “While the overall decline in mental well-being in younger ages is strongly present for males and females in the U.S., it is only present for females in India and not in males (where only select aspects deteriorated, while others improved). Even for females, it (the overall decline in mental well-being) is not as steep in India,” Dr. Thiagarajan said.

“On the other hand, both adolescent males and females in India have worse mental well-being on the whole, than their counterparts in the U.S. While aggression, anger and hallucinations are consistently related to the age of smartphone initiation for both U.S. and Indian females, for girls in India, getting their phones very young is more likely to result in increased sleep and health problems as adults,” she said.

Merits of ed-tech

In an attempt to address this, there is a growing debate on the merits of educational technology (ed-tech) in the elementary and middle school years. “One of the possible solutions is also to provide restricted access to phones for teenagers using apps, which lock in parental controls regarding apps teens can access, while allowing them to access a school portal or messaging,” Dr. Thiagarajan said.

Delhi HC lacks jurisdiction to hear copyright suit: OpenAI

Delhi HC lacks jurisdiction to hear copyright suit: OpenAI

TIMES NEWS NETWORK 24.01.2025

New Delhi : Days before a court hearing, artificial intelligence company OpenAI, which created ChatGPT, raised objections to the jurisdiction of Delhi HC to hear a copyright violation suit against the tech giant. In a submission, the Microsoftbacked company argued that Delhi HC lacks jurisdiction as OpenAI has no presence in the country, being situated primarily in the US. The company was responding to a suit filed by news agency ANI demanding that all of its data be removed from ChatGPT’s training modules. OpenAI contended that complying with ANI’s demand to remove its data would violate its legal obligations in the US. It is under a legal obligation, under the laws of the US, to preserve and not delete the said training data, as per a report. The company pointed out that it doesn’t have any permanent establishment in India and that its servers are outside the country. 

Last year, the news agency filed the suit against the tech firm, alleging it used ANI’s content without permission to train ChatGPT. In a hearing in Nov, OpenAI assured the HC it would not use ANI’s content anymore, but the news agency contended its published works were stored in ChatGPT’s servers and must also be deleted. It said the firm uses ANI’s published content for training its AI chatbot, ChatGPT, and further accused OpenAI of attributing fabricated news stories to the publication. ANI alleged that OpenAI “refused to obtain a lawful license or permission” to use its original works while it has entered into licensing agreements with news firms like The Financial Times and Associated Press for the use of their copyrighted content. During the hearing, HC had sought a detailed response from OpenAI to allegations that firm halt its use of ANI content in future AI model training. The court is scheduled to hear the matter on Jan 28

Teachers enhance their skills using an app to make classrooms interesting

Teachers enhance their skills using an app to make classrooms interesting

Several educators are using The Teacher App to stay updated with new teaching ideas

Priyadarshini.Gupta@timesofindia.com 20.01.2025

Designed to address challenges faced by educators, The Teacher App was recently launched by the Ministry of Education, which offers 900 hours of resources, including lesson plans, worksheets, project-based activities, and question banks for teachers. The app aims to evolve classroom engagement by supporting continuous training and application of skills to teachers. The app is a user-centric platform addressing the evolving needs of educators from government and private schools, B Ed students, and trainers. 

Designed by Bharti Airtel Foundation, the app allows teachers to frequently engage with modules addressing practical classroom challenges, subject-specific pedagogy, classroom management strategies, innovative Teaching Learning Materials (TLMs), digital teaching techniques, specially designed life skills courses, foundational literacy and numeracy (FLN) courses. The content has been curated for teachers and can be used by them at their own pace. The app was first introduced to the teacher community one and a half years ago for trials and gather feedback. 

Following a successful pilot in September, the app was officially launched in November last year. The launch saw an additional 12,500 teachers using the app. Approximately, 44,000 teachers are currently active users with over 35% of the registered teachers being active monthly users. Sumathi, a primary schoolteacher in Bengaluru, says she has utilised several teaching resources to enhance her teaching style. “Through the ‘Teaching Techniques’ course, I learned how to increase the ability of students to grasp concepts using engaging activities. For example, while teaching shapes, I encouraged children to form shapes like squares or triangles using their bodies in group activities. This hands-on approach helped students learn the concepts effectively who have shown signs of improved learning.”