Eric Schmidt: True AI Power Lies Beyond Language and Chatbots

Most people look at artificial intelligence and stop at chatbots. Words, sentences, clever answers. That’s where the fascination ends. But Eric Schmidt, who once ran Google, sees something else moving underneath, something harder to notice and maybe more important.

He recently spoke at a TED event, and what he pointed to wasn’t flashy. Not language. Not summaries. Planning. Strategy. That’s the shift. While the crowd is watching the show, he says, the real tech is learning how to think a few steps ahead—and do it alone.

Image: Ted / YT

Systems built on reinforcement learning, a method that lets them try, fail, adjust—are getting sharper. This isn’t about chat anymore. It’s about machines figuring out how to act without being told what comes next.

Right now, he says, everyone still thinks in terms of text. But AI is already going further. First it was language. Then came sequence—useful in biology, for instance. Now it’s about forming plans, solving problems in layers. And soon? Machines running whole business operations on their own. Quietly.

It’s Already in Use—Just Not Where Most People Are Looking

While people play around with bots that write poems or emails, deeper tools are being built. Schmidt mentioned a few: o3, Deepseek R1. They don’t just give answers. They try things out. Go back. Rethink. They loop.

• Also read: ChatGPT Statistics: Active Users Data Worldwide

He’s using some of them, personally. After investing in the space industry, he wanted to understand it better. Not casually—deeply. Instead of reading textbooks, he asked AI systems to study for him. One spent about 15 minutes pulling together a dense, research-heavy writeup. Schmidt called it “extraordinary.” Not polished, but insightful. Not written for style, but for depth.

Experts Say Planning Is Where Intelligence Starts to Look Real

Schmidt isn’t alone here. Yann LeCun from Meta has said similar things—today’s large language models don’t really think, he argues. They don’t understand space, memory, consequences.

To fix that, LeCun proposed a new kind of model—H-JEPA. Its goal? Let machines figure out steps, not just sentences. The idea is that, to be smart, an AI has to work with goals. It needs to try, adjust, move forward. That’s different from just guessing the next word.

And Schmidt agrees. Without that kind of reasoning, it’s all just smoke.

The Stakes Go Beyond Tech—It’s Geopolitical Now

He also framed the issue in terms of global power. Countries aren’t watching from the sidelines. The U.S. and China, he warned, are in a kind of AI arms race.

Trade restrictions have already changed how China approaches AI. With limited access to advanced chips, they’ve been pushed to rely on algorithmic workarounds—code over hardware. In a way, that’s making them faster.

Schmidt claimed that discussions about AI are happening inside military circles—real ones. Strategic ones. He even mentioned that in some rooms, people have suggested preemptive moves. That’s not science fiction. He thinks it’s five years away, maybe less.

Still, He Sees Room for Hope—If People Keep Up

Despite the tension, Schmidt isn’t all doom. He still sees AI as something that could change lives—in education, medicine, material science. He talked about AI tutors that adapt to the person, tools that help researchers explore what’s still unknown.

But he warned: this isn’t something to watch passively. “Ride the wave,” he said. Not once. Every day. The idea is, if you’re not using it, someone else is—and they’ll move faster than you.

In his mind, this isn’t like electricity or the internet. It’s bigger. AI, especially general intelligence, will shape the next century—or millennium. And it’s already starting to.



Read next: Future ChatGPT Could Store and Analyze Your Entire Digital Life
by Asim BN via Digital Information World

New Threads Update Makes Link Sharing More Powerful for Creators

For users posting on Threads — especially digital creators, media professionals, and online publishers — links have long felt like an afterthought. Posts containing URLs often struggled to find traction in the app’s recommendation engine, leaving many to wonder whether external content was being subtly discouraged.

This wasn’t entirely speculative. Instagram and Threads executive Adam Mosseri has openly acknowledged that links aren’t a priority in the platform’s content ranking process. While he stopped short of calling it intentional suppression, the outcome remained the same: posts with links tended to fade into the background.

That’s beginning to shift.

Meta, now pivoting harder toward creators as a cornerstone of Threads’ growth, is starting to adjust how links function on the platform. One of the more visible changes is a new option allowing users to embed up to five links directly within their bio — offering more flexibility than the previous single-link constraint.

More consequential, though, is how the app is starting to treat links in posts. Rather than fading out, posts that contain URLs are now being included more frequently in the app’s recommendations. And to make that visibility measurable, Meta has introduced new link-specific analytics that show creators how their audience engages with shared content — including how many taps a given link receives.

Meta, in a recent update, emphasized that the goal is to help creators expand their reach, even beyond the confines of the Threads platform itself.

While this may be welcome news for those hoping to convert their Threads presence into tangible traffic, questions remain. The recommendation engine still operates with limited transparency, and even high-follower accounts often see more reach from suggested posts than from their own followers. That dynamic makes the algorithm’s behavior harder to predict — especially when the performance of link-based content is on the line.

Media organizations and independent publishers, in particular, have reported inconsistent outcomes. A very small set of big publishers saw more reliable referral traffic from new networks like Bluesky, which despite its size, delivered steadier results. However, a recent uptick in link-driven engagement on Threads has been observed — especially after Meta rolled back its previous stance on promoting political content.

That said, caution remains warranted. Meta’s history is marked by frequent strategic pivots that have, in the past, disrupted the ecosystems of creators and publishers. Even so, with the platform loosening its grip on link visibility and engagement, there may be an emerging window for experimentation — one that could be worth exploring for anyone seeking new channels of audience growth.

Read next:

• Privacy Group Noyb Challenges Meta’s AI Training Plans Citing GDPR Violations

• ChatGPT Usage Statistics: Numbers Behind Its Worldwide Growth and Reach
by Irfan Ahmad via Digital Information World

ChatGPT Usage Statistics: Numbers Behind Its Worldwide Growth and Reach

OpenAI's ChatGPT has become an important part of people's lives, and as of May 2025, it has 170 million daily visitors (based on our estimates), with 800 million weekly active users. People rely a lot on ChatGPT, and it handles over 1 billion queries every day. With the rapid growth, it is expected that OpenAI will earn up to $11 billion in revenue by the end of 2025.

Total Number of ChatGPT users

In February 2025, ChatGPT had 400 million weekly users, which doubled to 800 million weekly users in March 2025. Open AI’s CEO, Sam Altman, also said during TED 2025 that ChatGPT’s user base doubled in just a few weeks.

In comparison, ChatGPT had just 50 million active weekly users in January 2023, which jumped to 100 million in August 2023. By October 2024, ChatGPT had 250 million weekly active users, which reached 300 million in December 2024.

Monthly Active Users (MAUs) of ChatGPT

In January 2023, ChatGPT had 100 million monthly active users, and it reached 173 million monthly active users in April 2023. In 2024, ChatGPT reached 180.5 million monthly active users. It reached 400 million weekly active users as of February 2025. According to a court document, revealed by Google, ChatGPT had approximately 600 million monthly active users in March. It should also be kept in mind that ChatGPT reached 100 million monthly active users just after two months of its launch, showing how quickly people started using it.

Daily Active Users (DAUs) of ChatGPT

In 2025, ChatGPT had an average of 170 million daily visitors. It's important to note that this visit count, calculated by Digital Information World using Similarweb’s April 2025 data, which reflects total visits, not unique users. As one person may visit multiple times, so actual active daily user numbers may vary. The daily average of 170 million visits was derived by dividing 5.1 billion monthly visits by 30 days. Most people use ChatGPT regularly, so its daily active users typically range from 148 to 200 million, based on data from tools like SEMRush.

How Many Queries Does ChatGPT Receive per Day?

According to data from OpenAI, ChatGPT handles over 1 billion queries every day. People heavily rely on ChatGPT for their questions, daily tasks, and for conversations.

Within a week of its launch, ChatGPT was already handling 10 million queries every day, and its usage has grown rapidly, as well as its user base.

OpenAI has also launched ChatGPT Pro on December 5th, which is priced at $200 a month and is best for engineers, researchers, and power users who need advanced AI usage daily.

User Demographics of ChatGPT

Gender:

There is a balanced gender split among users of ChatGPT, with 54% of its users being male and 46% of them being female, as per SimilarWeb insights.

Age:

Most of the users of ChatGPT are young adults and professionals. Around 55% of users of ChatGPT are between the ages of 18-34. Users aged 34–54 make up 32% of ChatGPT's total user base, while those aged 55 and older account for just 13%.

ChatGPT Users by Country

Even though ChatGPT is used globally, some countries have higher numbers of its users. The US has the highest number of ChatGPT users (18.11%), followed by India (7.99%) and Brazil (5.4%).

As per Similarweb, the UK has 3.39% of the total ChatGPT users, while Republic of Korea has 3.94% of users. 61.16% of the ChatGPT users are from different other countries.

ChatGPT Traffic Stats

As per SimilerWeb data, ChatGPT web traffic has grown a lot, especially in 2025. In April 2025, ChatGPT’s website got 5.1 billion visits, which is a 30.77% increase from 3.9 billion in February 2025. This shows that the engagement with ChatGPT is rising. Between late 2023 to early 2025, traffic on ChatGPT was steady, and jumped between September and October 2024 from 3.1 billion to 3.7 billion. In April 2024, there were 1.81 billion monthly visits on ChatGPT, and it reached 5.1 billion monthly visits in April 2025.

Month	Monthly ChatGPT Website Visits
April 2025	5.1 billion
March 2025	4.5 billion
February 2025	3.9 billion
January 2025	3.8 billion
December 2024	3.7 billion
November 2024	3.8 billion
October 2024	3.7 billion
September 2024	3.1 billion
August 2024	2.6 billion
July 2024	2.44 billion
April 2024	1.81 billion
January 2024	1.6 billion
December 2023	1.6 billion
November 2023	1.7 billion
February 2023	1 billion
January 2023	616 million
December 2022	266 million
November 2022	152.7 million

Revenue of ChatGPT

OpenAI, which is the parent company of ChatGPT, is getting a lot of revenue from it. According to the CFO of OpenAI, Sarah Friar, ChatGPT is going to reach $11 billion in revenue in 2025. In 2022, ChatGPT had less than $10 million in revenue, and it reached $1 billion in 2023. In 2024, the revenue of ChatGPT was $3.7 billion. That means OpenAI’s revenue reached around $10 billion within three years, which shows how much demand there is for ChatGPT globally.

ChatGPT Plus Users

There are around 11 million users of ChatGPT who are subscribed to ChatGPT Plus, with around 1 million users subscribed to team and business plans, which give them more advanced features.

Average User Session

On average, users spend 13 minutes 38 seconds per session on ChatGPT, while 14 minutes 14 seconds was the longest session time recorded in October 2023.

ChatGPT Users as Compared to Other Platforms

Platform	Time To Reach 100 Million Users	Industry
Instagram Threads	2 days	Social Media
ChatGPT	2 months	Chatbot
TikTok	9 months	Social Media
Youtube	18 months	Social Media
Instagram	30 months	Social Media
Facebook	54 months	Social Media
Twitter	60 months	Social Media
Spotify	132 months	Music Streaming
Netflix	216 months	Entertainment

ChatGPT has grown faster than any other major tech platform, reaching 100 million users in just 2 months. Only Instagram Threads was faster than ChatGPT in reaching that milestone in just 2 days, but its usage has dropped while ChatGPT’s usage continues to increase.

The third fastest app to reach 100 million users was TikTok, and did so in 9 months, while YouTube reached 100 million users in 18 months after its launch

ChatGPT also reached 1 million users in just 5 days, the fastest to do so after Instagram Threads, which did so in just an hour. Instagram reached 1 million users in 2 months, while Spotify reached this milestone in just 5 months.

Platform	Year of Launch	Time To Reach 1 Million Users
Instagram Threads	2023	1 hour
ChatGPT	2022	5 days
Instagram.	2010	2 months
Spotify	2008	5 months
Dropbox	2008	7 months
Facebook	2004	10 months
FourSquare	2009	13 months
Twitter	2006	2 years
Airbnb	2008	2.5 years
Kickstarter	2009	2.5 years
Netflix	1999	3.5 years

Interesting Facts:

As per SimilarWeb data 61.05% of ChatGPT's traffic comes from YouTube, which means from all the social media platforms YouTube viewers are the largest referral source of its user base, likely driven by content creators, tutorials, and AI-related videos linking to the platform.

According to SimilarWeb data, the United States was the biggest market for ChatGPT in April 2025, accounting for approximately 924 million of the 5.1 billion visits, or 18.11% of the total traffic.

• Also read: Generative AI Platforms See Remarkable Engagement, With Users Spending 6+ Minutes per Session

Data from SEMrush suggests that many users still either mispronounce or mistype the popular term "ChatGPT," with some searching for "Chat GBT" (673,000 monthly searches as of April 2025) and "Chat Got" (246,000 searches), instead of ChatGPT.

Market Share of ChatGPT

Description	AI Search Market Share	Quarterly User Growth (Est.)
ChatGPT (excluding Copilot)	GPT-3.5, GPT-4	59.90%
Microsoft Copilot	GPT-4	14.30%
Google Gemini	Gemini	13.40%
Perplexity	Mistral 7B, Llama 2	6.30%
Claude AI	Claude 3	3.30%
Grok	Grok 2, Grok 3	0.70%
Deepseek	DeepSeek V3	0.70%
Komo	Not publicly disclosed	0.60%
Brave Leo AI	Mixtral 8x7B	0.30%
Andi	Not publicly disclosed	0.20%

ChatGPT leads in market share among the generative AI chatbot space, with 59.2% of the total AI search market. It is followed by a 14.4% share of Microsoft CoPilot and a 13.5% share of Google Gemini.

ChatGPT App Statistics

Overall, ChatGPT has 34 million average monthly downloads and is rated 4.5 stars on Google Play and 4.9 stars on the Apple App Store. On the App Store, ChatGPT is also ranked number 1 in productivity apps.

In January 2025, ChatGPT got 37 million new downloads, a 4.02% decrease from 38.6 million in December 2024. The highest number of downloads of ChatGPT recorded was in April 2025 (52 million).

Conclusion:

ChatGPT is seeing rapid growth, with users increasing monthly in large amounts. It is because ChatGPT offers the best AI technology and responses, and gives users a platform for different purposes.

Read next: 

• Offline Nations in an Online Age: Mapping the World's Digital Divide

• Chatbots Hallucinate More With Confident or Short Prompts, Accuracy Drops Up to 20% in Critical Tasks

• OpenAI Becomes the Default Setting for Corporate AI Spend

• US Data Centers Projected to Consume 606 TWh of Energy by 2030
by Arooj Ahmed via Digital Information World

OpenAI Unveils Safety Tracking Hub Amid Transparency Concerns and Legal Pressure

Tech giant OpenAI introduces a much-needed safety evaluations page, meant to track how its models behave when pushed beyond its limits. Rather than waiting for users questions to pile up, the company now puts confusion patterns, bad answers, obedience gaps, and trick responses together under one roof.

It is important to note that the launch of this Safety Evaluations hub didn’t come out of the blue. OpenAI has been under fire lately. Multiple lawsuits claim it has relied on protected material to train its systems.

This new safety hub expands earlier efforts. In the past, system cards gave one-time reports when a model launched. Those weren’t updated often. This new hub, however, should evolve over time. It includes performance details about GPT-4.1 up through 4.5 and keeps that data open to visitors.

Though it sounds useful, the page isn’t flawless. OpenAI checks its own work. It also decides what gets shared. That makes it harder for outsiders to trust everything shown there. Which means there’s no third-party audit, no independent voice checking what’s missing or misrepresented.

OpenAI says it wants better visibility into how its models perform. But it holds the steering wheel and the map. So, while the platform may bring progress, it still leaves observers wondering what they’re not seeing.

Image: DIW-AIgen

Read next: Banned Without Warning: Pinterest Apologizes Late, Users Still Distrust Platform
by Irfan Ahmad via Digital Information World

Banned Without Warning: Pinterest Apologizes Late, Users Still Distrust Platform

For weeks, Pinterest users were left in the dark — locked out of their accounts, confused, and, in many cases, furious. Without any notice, users found their profiles suspended or content suddenly gone. Many of them insisted they had followed the rules. But still, the bans kept happening. And during all that time, Pinterest said almost nothing.

People turned to Reddit, X, and community forums, trying to figure out what was going on. Some had lost years of saved Pins, carefully collected over time. Others said their normal posts had vanished without explanation. When they reached out to support, they got cold or copy-paste replies—if they heard back at all.

That silence only made things worse.

At first, when Pinterest did speak up, on May 1, it didn’t say much. The platform simply asked affected users to send private messages, as if the issue wasn’t widespread. There was no clear apology, no public plan, and for many users, no comfort. Some began talking about possible legal steps or even messaging Pinterest’s executives directly on LinkedIn.

But now, finally, there’s some clarity.

On May 13, Pinterest officially admitted there had been a mistake. The company said the bans weren’t caused by any AI system, as many had assumed. Instead, the issue came from a problem inside their own systems — and some accounts had been flagged and blocked by accident.

Pinterest said it’s already restoring access for users who were wrongly banned. It also promised to improve how these kinds of errors are handled in the future.

Still, for many users, the apology came too late. Trust has been shaken. People are saying the damage is already done. And although the company has started making things right, some feel they were left unheard for too long.

This issue with Pinterest isn’t a one-off. It reflects a broader problem across the entire social media industry. Many platforms, including Meta, have leaned too heavily on automated systems and artificial intelligence for content moderation and account verification. In Meta’s case, some users are being asked to verify their identity using a video selfie, a process largely controlled by AI. But instead of improving safety, this tech-driven approach often ends up rejecting real people, locking them out for no valid reason, and offering no clear way to appeal. It’s a growing trend: less human support, more machine errors, and a worse experience for the very users these platforms are built for.

Image: DIW-Aigen

Read next:

• Generative AI Platforms See Remarkable Engagement, With Users Spending 6+ Minutes per Session

• Tools of the Future, Shame of the Present: Why AI Users Stay Quiet at Work

• The World's Largest Unconnected Populations
by Irfan Ahmad via Digital Information World

Offline Nations in an Online Age: Mapping the World's Digital Divide

While the world scrolls nonstop, millions of people remain outside the digital map — entirely offline.

A fresh April 2025 analysis by We Are Social and Meltwater revealed the scale, showing that India alone has over 600 million people still not using the internet. That’s more than the entire population of Europe combined. Even with growing telecom reach, India’s offline share stood around 41 percent. China follows far behind with 303 million unconnected people, though its relative disconnection rate is only about 21 percent, thanks to stronger urban digital infrastructure. Pakistan ranks third in terms of offline population, with over 138 million individuals still cut off from the web, which forms a shocking 54 percent of its entire population.

Location	Unconnected Individuals	% of Pop. Offline
India	606,997,000	41.60%
China	303,220,000	21.40%
Pakistan	138,391,000	54.40%
Nigeria	128,337,000	54.30%
Ethiopia	105,472,000	78.40%
Bangladesh	93,251,000	53.20%
Dem. Rep. of the Congo	77,738,000	69.50%
Indonesia	61,911,000	21.70%
Tanzania	49,685,000	70.90%
Uganda	36,647,000	71.80%

In absolute terms, Nigeria and Ethiopia also show worrying figures. Nigeria has 128 million people not online, while Ethiopia, with fewer total people, still sees 105 million citizens offline — making up nearly four-fifths of the country. In Bangladesh, the digital gap touches over 93 million people, representing more than half of its population. The Democratic Republic of the Congo trails behind that, but 69 percent of its people still remain without digital access. Indonesia, despite major strides, still holds over 61 million people offline, while Tanzania and Uganda show rates near or above 70 percent, placing them among the highest in percentage terms.

When the data is arranged by proportion of population rather than total heads, the map shifts. North Korea, not surprisingly, leads with over 99 percent of its population offline, although exact figures remain blocked or hidden from public view. After that, the pattern sharpens across smaller African nations. Burundi and Chad report over 86 percent of citizens without internet. In Central African Republic and South Sudan, the unconnected population hits 84 percent. Yemen, where conflict has strained infrastructure, leaves over 34 million people offline, forming 82 percent of its total population. Malawi, Mozambique, Eritrea, and Madagascar all report digital disconnection rates above 79 percent, despite global shifts toward mobile-first solutions.

Digital Isolation in 2025: Top Countries With the Lowest Internet Penetration Rates:

Location	% of Pop. Offline	Unconnected Individuals
North Korea	>99%	[Blocked]
Burundi	87.20%	12,473,000
Chad	86.80%	18,115,000
Central African Republic	84.20%	4,604,000
South Sudan	84.10%	10,195,000
Yemen	82.30%	34,136,000
Malawi	82.00%	18,104,000
Mozambique	80.20%	28,361,000
Eritrea	80.00%	2,870,000
Madagascar	79.60%	25,917,000

Some countries appear in both categories — large numbers and high percentages — such as Ethiopia and Bangladesh. But others offer contrast. Indonesia has more unconnected people than Eritrea, but its population size masks the percentage gap. Likewise, Uganda and Tanzania report smaller populations, yet their percentage of people offline remains troublingly high. In nations like Mozambique and Madagascar, both indicators rank poorly, highlighting structural issues that go deeper than affordability or access.

Experts usually link high disconnection rates to a mix of economic limitations, infrastructure gaps, and sometimes political restrictions. In many cases, education, employment, and healthcare are all tied to basic internet access. While mobile internet growth has shifted the dynamic in cities, rural areas continue to fall behind. The digital divide, though shrinking in global averages, is still sharply visible when broken down by country.

These figures come from multiple sources including the ITU, GSMA Intelligence, and official national reporting. While they reflect the most recent updates available, regional data gaps or late publications may cause slight shifts. Even so, the April 2025 picture confirms a truth the web often forgets — not everyone is online, and progress is far from evenly spread.

Source: Datareportal

Read next:

• Generative AI Platforms See Remarkable Engagement, With Users Spending 6+ Minutes per Session

• Study Reveals How Google’s AI Overviews Change Search Behavior and User Trust
by Irfan Ahmad via Digital Information World

Android 16 Brings Major Security Upgrades and Smarter Anti-Theft Features

Android 16, the next major update for Android devices, is set to bring important security enhancements this summer. Google aims to combat fraud and prevent the exploitation of stolen phones with several new features. After introducing AI-driven tools in March to spot scam messages using common text patterns, Google is now expanding its approach. The tech giant is training its models to detect a wider range of fraud, including road toll scams, which have become more prevalent this year. Plus, Google Play Protect will now be quicker at identifying harmful apps.

One of the standout changes in Android 16 is the restriction on granting certain accessibility privileges during calls with unknown numbers. This is paired with the introduction of the Key Verifier feature, which prevents scammers from impersonating your contacts. The new tool allows users to exchange public encryption keys with trusted people, making it easier to verify each other's identities.

For users concerned about phone theft, Android 16 adds more layers of protection. A new feature, Identity Check, now requires biometric verification to alter sensitive settings if you're outside of a trusted location. Originally launched on certain Pixel and Galaxy phones in January, it will now be more widely available. This aims to prevent incidents where thieves steal phones by observing passcode entry in public spaces, a common issue in crowded places like bars.

Other anti-theft measures include a new security question when trying to disable remote locks, along with stricter limitations on actions taken after a factory reset. Android 16 will also hide two-factor authentication codes received via text on the lock screen if the phone has been offline or not unlocked for some time.

The update also introduces Advanced Protection, a centralized security feature that groups multiple settings together. Once activated, no part of the system under this umbrella can be turned off.

Google is also enhancing its Find My Device tool by expanding it into Find Hub. This new service allows users to track any object with a Bluetooth tag or smart technology. Devices equipped with ultra-wideband (UWB) will offer more precise location tracking, and satellite communication ensures functionality even outside cellular network coverage.

These updates were revealed during Google’s pre-recorded Android Show on May 13, which also highlighted the introduction of Material 3 Expressive design, expanded Gemini AI for Wear OS and Android Auto, and additional security improvements. Google plans to make further announcements about AI developments at the Google I/O 2025 conference, taking place from May 20 to May 23.

Read next: Google Homepage Pushes AI Mode Forward, Leaves 'I'm Feeling Lucky' Button Behind
by Asim BN via Digital Information World