The company announced on Monday that it had submitted a confidential S-1 filing with the U.S. Securities and Exchange Commission (SEC), but said it had not yet decided when to go public.

In a statement, OpenAI said: “We recently submitted a confidential S-1. We expect it to leak so we’re just announcing it. We have not decided on timing yet; it may be a while because there are things we want to do that are likely easier as a private company. But it’s a complicated set of tradeoffs and this gives us the option to go public sooner if that ends up being best.”

The filing places OpenAI alongside competing AI developers Anthropic and Perplexity, which also submitted confidential IPO documents last week and this week.

Meanwhile, Elon Musk’s SpaceX is preparing for its own stock market debut, setting up what could become one of the biggest waves of technology listings in recent years.

Although OpenAI did not disclose the size of the proposed offering or a timeline for the listing, this development will boost the company’s growth since the launch of ChatGPT in 2022.

OpenAI has grown into one of the world’s most valuable private firms, with latest valuation standing at about $852 billion, while Anthropic was recently valued at nearly $965 billion following a new funding round.

According to reports, OpenAI has been working with investment banks Goldman Sachs and Morgan Stanley on the IPO process.

The company is also expected to organise a tender offer that would allow employees to sell some of their shares, providing liquidity ahead of any public listing.

OpenAI, however, is still facing challenges to justify its valuation as it spends heavily on computing power and infrastructure needed to train and operate more advanced AI models.

The company has reportedly raised more than $180 billion from investors but is still in a phase of significant spending.

Chief Executive Officer Sam Altman said in a recent blog post that the company is entering what he described as its “third phase”.

According to Altman, OpenAI first focused on research into artificial general intelligence before becoming a product company through services such as ChatGPT.

“Now we are entering the third phase,” Altman wrote. “The economy is beginning to reshape around AI. The central question now is how to make advanced AI abundant, affordable, safe, useful, and easy enough for every person and organization to benefit from it.”

The planned IPO arrives shortly after a case involving OpenAI and Musk. An advisory jury recently ruled against claims brought by Musk, who had accused OpenAI and Altman of abandoning commitments related to the company’s original non-profit mission. A federal judge subsequently adopted the jury’s verdict.

The number of AI companies preparing for public listings is also drawing attention across the wider industry.

AI search startup Aravind Srinivas said Perplexity still intends to pursue an IPO in 2028 regardless of how the listings of OpenAI and Anthropic perform.

“Agnostic of these two companies, we were planning for something in 2028, so that still remains the case,” Srinivas told CNBC.

He acknowledged that investor reaction to upcoming AI listings could influence market sentiment, particularly the reception to SpaceX’s offering.

“I certainly think there will be ripple effects if they don’t go well, like there is no sugar coating on that. The SpaceX IPO this week will definitely be a leading indicator of how Anthropic or OpenAI will go out,” he said.

Srinivas added that strong public market performances would benefit the AI sector.

“I think it’s important for the AI industry that these IPOs go well, and I actually think they will go well, because they’re doing well.”

The lawsuit, filed on March 13, 2026 at the U.S. District Court for the Southern District of New York, claims that OpenAI copied close to 100,000 articles from Britannica and its dictionary arm, Merriam-Webster.

According to the complaint, this material was used to train large language models behind ChatGPT.

Britannica says the chatbot can produce responses that are so close to its original entries. It argues that this reduces visits to its own platforms, as users rely on AI-generated summaries instead of going directly to its website.

The company also alleges trademark misuse, stating that its name is sometimes cited in responses in ways that suggest approval where none exists.

In response, OpenAI said its systems are built using publicly available data and operate within the bounds of fair use. A spokesperson said, “Our models empower innovation, and are trained on publicly available data and grounded in fair use.”

Britannica has not publicly expanded on the case since filing. Its representatives did not respond to follow-up questions at the start of the week.

This case adds to other legal disputes involving AI developers and content owners. Over the past year, publishers, authors and artists have challenged how their work is used in training data.

Similar claims have also been brought against other firms, including an ongoing case between Britannica and Perplexity AI, filed in 2025.

Questions including whether using copyrighted material to train AI systems qualify as fair use, have been asked. Technology companies say the process transforms the material into something new, but Publishers disagree, arguing it amounts to copying without consent.

Microsoft, Meta, Amazon, Perplexity, and Mistral AI have all joined Google in paying the Wikimedia Foundation for structured access to its projects, including the online encyclopedia’s 65 million articles.

These agreements are part of Wikimedia Enterprise, a paid API launched in 2021 that provides high-quality, reliable data tailored for large-scale use by companies. 

The initiative allows tech firms to tap Wikipedia’s content without overwhelming its servers, a problem that had been growing as companies relied more on freely available articles for AI model training.

“Wikipedia is a critical component of these tech companies’ work that they need to figure out how to support financially,” Lane Becker, president of Wikimedia Enterprise, told Reuters. 

“It took us a little while to understand the right set of features and functionality to offer if we’re going to move these companies from our free platform to a commercial platform … but all our Big Tech partners really see the need for them to commit to sustaining Wikipedia’s work.”

The timing of the announcement coincides with Wikipedia’s 25th anniversary, highlighting how far the volunteer-driven platform has come since its launch. 

What began as a community-powered encyclopedia is now a vital resource for the global tech industry, underpinning much of the data used in AI development.

Microsoft’s Tim Frank, corporate vice president, noted the importance of the development: “Access to high‑quality, trustworthy information is at the heart of how we think about the future of AI at Microsoft … (With Wikimedia), we’re helping create a sustainable content ecosystem for the AI internet, where contributors are valued.”

Traditionally funded by small public donations, Wikipedia has faced increased server costs due to large-scale content use by AI developers. 

The Enterprise programme addresses this by providing structured, paid access, creating a sustainable revenue model without undermining the free encyclopedia that millions rely on.

Adding to this era of change, Bernadette Meehan, former U.S. Ambassador to Chile, took over as Wikimedia Foundation CEO on January 20, 2026. 

Her appointment is a professionalisation of leadership as the organisation balances public service with commercial partnerships.

The new agreements formalise Wikipedia’s function in the data supply chain for tech giants while protecting the volunteer work that underpins the platform.

This immediately sent the company’s shares up by 16% after trading hours.

Under the one-year deal, paid in cash and equity, Perplexity’s AI-powered “answer engine” will become part of Snapchat’s chat interface, letting users ask questions and receive verified, real-time responses without leaving the app. 

Revenue contributions from the partnership are expected to begin in 2026.

Snap CEO Evan Spiegel confirmed that the new feature will be ad-free. “Perplexity AI will control the responses from their chatbot inside of Snapchat. So, we won’t be selling advertising against the Perplexity responses,” he said.

The collaboration will enable Snap to compete more aggressively with social media giants like TikTok and Meta’s Facebook and Instagram, both of which have long topped the advertising market. 

Analysts see mutual benefits as Perplexity gains exposure among younger audiences, while Snap strengthens user engagement within its platform.

“Perplexity needs a way to build its profile among young consumers, and Snap needs an AI chat partner that will allow its users to stay engaged without leaving its app,” said Max Willens, principal analyst at Emarketer.

Snap’s own AI tool, “My AI,” will be available, but users will now have the option to interact with Perplexity’s chatbot instead. Spiegel described the initiative as part of Snap’s goal to make artificial intelligence “more personal, social, and fun.” He also noted that more third-party AI integrations could follow.

The announcement came alongside Snap’s third-quarter earnings, which showed a 10% year-on-year revenue increase to $1.51 billion, beating analyst expectations of $1.49 billion, according to LSEG data. 

The company’s net loss narrowed to $104 million, down from $153 million a year earlier, while daily active users climbed 8% to 477 million globally.

Snap’s focus on direct-response advertising continued to pay off, with that segment rising 8% during the quarter. Growth was driven by demand for its “Pixel Purchase” and “App Purchase” tools, which help businesses reach users more likely to complete transactions.

Still, the company warned of possible challenges in the coming quarter. Snap expects daily active users could decline slightly due to new regulatory changes, including Australia’s Social Media Minimum Age Bill, which takes effect in December and will tighten access for younger users.

For the final quarter of 2025, Snap projected revenue between $1.68 billion and $1.71 billion, broadly in line with market expectations of $1.69 billion.

The outage, which originated from AWS’s US-EAST-1 region, began in the early hours of Monday and quickly spread beyond the United States, affecting Europe, Asia, and Africa. 

According to AWS’s own status dashboard, multiple services are currently “impacted” due to “increased error rates and latencies,” with engineers “actively engaged and working to both mitigate the issue and understand root cause.”

For users, the impact has been immediate and widespread. Alexa devices have gone silent, unable to respond to voice commands or execute daily routines like alarms and reminders. 

Developers and businesses using AWS’s cloud network, from Airtable to Perplexity AI and the McDonald’s app, have also been hit. Even high-traffic entertainment platforms like Fortnite, Roblox, and Rainbow Six Siege are offline.

Downdetector, a platform that tracks service disruptions, has logged over 2,000 incident reports in the U.S. alone since the outage began. On Reddit and X (formerly Twitter), frustrated users across time zones have shared screenshots of failed connections and frozen dashboards.

“Perplexity is down right now,” confirmed Aravind Srinivas, CEO of Perplexity, in a post on X. “The root cause is an AWS issue. We’re working on resolving it.”

Amazon, in its latest public update at 3:51 a.m. ET, noted that it would provide further information every 45 minutes “or sooner if we have additional information to share.” However, at the time of writing, there is still no estimated timeline for full restoration.

This isn’t the first time AWS’s US-EAST-1 region has been the source of widespread disruption. Similar outages in December 2021, November 2020, and June 2023 took down high-profile platforms including Netflix, Disney+, Slack, Zoom, and Twitch. 

Each incident revealed an issue across the tech industry, that a large portion of the global internet depends heavily on a single cloud provider’s regional infrastructure.

The current outage appears to have hit both consumer-facing apps and backend systems, including AWS’s own Support Center and Support API, which organisations rely on for case creation and troubleshooting.

While AWS has reiterated that engineers are investigating the problem, the lack of transparency about the specific cause of the outage is driving industry-wide anxiety. Many are now revisiting familiar cases of how much centralisation is too much when the internet’s backbone depends on just a handful of companies.

For now, millions of users are in a holding pattern, waiting, refreshing, and hoping their devices come back online soon.

In a letter addressed to Perplexity CEO Aravind Srinivas, the BBC accused the company of scraping its online material without consent and warned of possible legal consequences if its demands were not met.

Specifically, the BBC is asking for the deletion of all copies of its content used for training purposes, an end to the practice, and a proposal for financial compensation. Failure to comply, it says, could lead to an injunction.

This escalation places Perplexity at the centre of a deepening rift between traditional media outlets and tech companies leveraging journalistic content to power artificial intelligence systems.

“The BBC’s letter outlines a potential injunction unless Perplexity halts its content scraping activities, purges existing data from its models, and presents a proposal for financial compensation,” reported Financial Times, which had access to the communication.

Perplexity’s reaction was quick as it said in a statement also quoted by FT, “The BBC’s claims are manipulative and opportunistic,” adding that the broadcaster “has a fundamental misunderstanding of technology, the internet and intellectual property law.”

This isn’t the first time Perplexity has drawn the ire of publishers. In October 2024, The New York Times served the company with a cease-and-desist notice over similar allegations. The paper demanded a full stop to the use of its articles in AI training and sought answers on how Perplexity bypassed anti-scraping mechanisms.

Other major publishers such as Forbes, Wired, and Axel Springer have had similar complaints. While Perplexity insists it is indexing publicly accessible content rather than scraping it for training, companies are not convinced.

To manage the case, Perplexity launched a Publishers Program in mid-2024, offering revenue-sharing arrangements to selected media outlets.

Among the early participants are TIME, Der Spiegel, The Texas Tribune, and Fortune. These partners receive a share of ad revenue whenever their content appears in responses generated by the platform.

We don’t know yet if the BBC was ever approached to join the scheme—or declined.

The broadcaster’s demand is a defensive legal move which stresses the need for a transparent and enforceable licensing structure for how AI firms use journalistic content.

Perplexity, which is backed by high-profile investors including Amazon’s Jeff Bezos, continues to defend its operations. The company maintains it is not infringing on intellectual property rights and blames the issue on misinterpretations of how its system works.

In contrast, OpenAI’s GPT-4 generates 25 times more emissions per query, pointing out issues about the environmental cost of advanced AI models.

A recent analysis by TRG Datacenters compared the carbon footprints of various AI models, measuring emissions from individual queries based on standard energy grid assumptions. The results reveal a gap in efficiency, with some models demanding far more power than others.

The AI Carbon Footprint Breakdown

AI chatbots differ widely in how much energy they consume during inference. Here’s how they rank in terms of CO2 emissions per query:

• Grok AI – 0.17g
• Google Gemini – 1.6g
• LLaMA (Meta AI) – 3.2g
• Claude AI – 3.5g
• Perplexity AI – 4g
• ChatGPT (GPT-4) – 4.32g

Grok AI’s low emissions result from its simplified computational design, which reduces power usage while maintaining performance. In practical terms, a single query on Grok produces the same emissions as a basic Google search, making it the greenest option in the market.

Meanwhile, Google Gemini ranks second, emitting 1.6g of CO2 per query. Google’s heavy investment in renewable energy and custom AI hardware helps curb its carbon footprint, though it still lags behind Grok.

Meta’s LLaMA model follows at 3.2g CO2 per query, benefiting from Meta’s commitment to renewable energy but still consuming twice as much power as Gemini. Claude AI ranks slightly worse, producing 3.5g CO2 per query, with its emphasis on safety and reliability seemingly driving up computational costs.

At the higher end of the spectrum, Perplexity AI (4g CO2 per query) and ChatGPT (4.32g CO2 per query) stand out for their environmental impact. GPT-4, in particular, has the highest carbon footprint among the chatbots studied. 

Its computational intensity, deep learning architecture, and search feature demand massive energy resources, generating emissions equivalent to sending 21 emails or nearly a full phone charge per query.

A spokesperson from TRG Datacenters commented on the findings:

“As AI adoption continues to rise, finding ways to reduce its energy consumption will be key. Some models are already designed to be more efficient, but there is still room for improvement. Advances in hardware, more optimized AI models, and increased use of renewable energy in data centres could help lower emissions over time. AI is here to stay, but balancing innovation with sustainability will be essential in minimizing its environmental impact.”

With AI usage skyrocketing, energy efficiency is becoming a big issue. While Grok AI sets the benchmark for low-carbon performance, larger models like GPT-4 highlight the environmental trade-offs that come with high-powered AI capabilities.

The resilience of AI sustainability will probably depend on hardware improvements, algorithmic optimisations, and increased reliance on green energy. For now, the numbers show that not all AI models are created equal, and some come with a much heavier environmental cost than others.

Per Reuters, the AI-powered search startup, backed by investors including Amazon founder Jeff Bezos and SoftBank, is looking to scale its operations as demand for AI-driven search continues to grow.

The company, which provides real-time search results by scanning the internet, has quickly increased its market value. In mid-2024, Perplexity was valued at $3 billion, a figure that jumped to $9 billion by December. 

If the latest funding round materialises, the valuation will double again, placing Perplexity as one of the most valuable AI startups.

With the rise of AI-powered tools, Perplexity has been enhancing its platform to compete with Google’s Gemini and OpenAI’s ChatGPT. The startup recently teased Comet, a web browser designed to handle complex queries, automate tasks, and make independent decisions. 

Its Deep Research feature further strengthens its offering by scanning multiple sources to generate comprehensive reports.

Competition in the AI search sector is increasing. OpenAI has integrated search into ChatGPT, while Google has rolled out AI Overviews within its search engine—though not without controversy due to factual errors. Anthropic, another AI company, has also introduced a search-enabled chatbot, increasing the pressure on Perplexity to differentiate itself.

Nonetheless, Perplexity has been warned by news organisations, including Forbes and Wired, for allegedly repurposing content without proper attribution. 

In response, the company introduced a revenue-sharing model, pledging to compensate publishers when their content is cited in AI-generated responses that generate ad revenue.

The device, which will be priced under $1,000, is set for release in the second half of the year, with sales expected to begin in 2026.

Claudia Nemat, DT’s board member in charge of technology and innovation, emphasised that while the company is not building large language models, it is focusing on AI agents. “We are becoming an AI company,” she stated during a press conference at the Mobile World Congress (MWC) in Barcelona.

Perplexity, known for its AI-powered search engine, will be a huge part of the phone’s AI features. The company’s CEO, Aravind Srinivas, explained its approach: “Perplexity is transitioning from just being an answer machine to an action machine. It is going to start doing things for you, not just answering questions. It’s going to be able to book flights for you, book reservations for you, send emails for you, send messages, place phone calls for you, and all those sorts of things, like set smart reminders.”

Although specific details about the device’s hardware and operating system remain undisclosed, Nemat noted that the phone would incorporate AI features directly into its core functionality, including lock screen interactions. It will also integrate services from Google Cloud AI, ElevenLabs, and Picsart.

T-Mobile’s AI assistant, Magenta AI, will be available as an app for Android and iOS users, but only for existing T-Mobile customers. 

This launch aligns with telecom companies’ focus on regaining influence in the smartphone sector, which has long been filled up by manufacturers and software giants like Apple and Google.

Perplexity’s involvement in the project comes as it seeks to differentiate itself in the competitive AI space, where companies like OpenAI, Anthropic, and Google continue to push boundaries. “These are the kinds of things that earlier you would have to do in your own way, learning how to use these different apps,” Srinivas noted. 

“All these things are going to start becoming easier so that you can focus your time and energy on problem-solving.”

Telecom providers are looking to leverage new technology to strengthen their market position and T-Mobile seeks to disrupt the smartphone market.

The majority of the funding will come from external investors, but Perplexity itself will also contribute, using some of the capital it previously raised for its own expansion.

The fund will be managed by Kelly Graziadei and Joanna Lee Shevelenko, who previously co-founded the venture firm F7 Ventures. It’s still unknown if they will continue their involvement with F7 or shift their full focus to Perplexity’s investment initiative.

This comes shortly after Perplexity secured $500 million in funding last December, tripling its valuation to $9 billion. The investment round included backing from SoftBank, Nvidia, and Amazon founder Jeff Bezos.

Perplexity is not the only AI company with a dedicated investment fund. OpenAI also runs the OpenAI Startup Fund but has stated that it does not invest its own money, unlike Perplexity.

Added to its investment, Perplexity has been expanding its product lineup. The AI search engine developer recently introduced ‘Deep Research,’ a feature for generating in-depth AI-driven reports, and is developing a web browser named Comet, for which users can now join a waitlist.

The AI sector is attracting high venture capital. A report from HSBC Innovation Banking revealed that 42% of all U.S. venture capital investments in 2024 went into AI startups, compared to 36% in 2023 and 22% in 2022.

“Venture capital has always gravitated toward transformative industries, but the level of consolidation we’re seeing within one category is unprecedented,” said Dave Sabow, gead of HSBC U.S. Innovation Banking. 

“The radical change this investment will fuel places us in the dawn of ‘The Agentic Age,’ an era where autonomous artificial intelligence capabilities fundamentally redefine how we communicate, work and interface with digital and physical worlds.”

In recent months, AI and robotics startups have been raising funds. Figure, a company developing humanoid robots, is reportedly in talks to secure $1.5 billion in Series C funding.

Lambda, a cloud AI firm, has raised $480 million, while Together.ai, which focuses on open-source AI models, has secured $305 million in Series B funding.