The University of Chicago Law School will prohibit incoming first-year students from using laptops and mobile phones in required courses as part of a sweeping new artificial intelligence policy designed to strengthen foundational legal reasoning before students begin using AI tools in their legal education.

The policy, unveiled on Thursday, is among the most restrictive AI frameworks adopted by a U.S. law school and reflects a growing debate within legal education over how to balance the benefits of generative AI with the need to preserve core analytical and advocacy skills.

According to the law school, the restrictions are intended to ensure that students “actually learn to think critically, strategically, and independently without relying on AI” before they are taught how to integrate the technology into legal practice.

Dean Adam Chilton said he was unaware of any other U.S. law school that has imposed a blanket ban on laptops and phones across all required first-year courses.

Ban Targets Foundational Legal Training

The new rules will apply to all mandatory first-year subjects, including constitutional law, contracts, torts, civil procedure, criminal law and property, where students traditionally develop the analytical framework that underpins legal education.

Under the policy, students will not be permitted to use laptops or mobile phones during class. Instead, professors may appoint designated “scribes” who will use electronic devices to take shared notes that can later be distributed to classmates.

The law school also plans to administer all examinations for required first-year courses in person without access to the internet, electronic documents, AI tools or other digital applications. The measures will initially be introduced on a pilot basis during the upcoming academic year.

The policy comes as generative AI tools such as ChatGPT, Claude and Gemini become increasingly capable of summarizing cases, drafting legal arguments, conducting legal research and generating written submissions.

While many law firms have begun integrating AI into their operations to improve productivity, legal educators remain divided over how extensively students should rely on the technology during training.

Chilton said the school’s objective is not to reject AI but to ensure students first master the reasoning skills that lawyers need when technology is unavailable or inappropriate.

“AI is forcing us to ask ourselves, ‘What are the essentially human skills that we should be training that AI can’t replace?'” Chilton told Reuters.

He said lawyers are frequently required to answer judges’ questions during court proceedings, advise clients in real time, or negotiate with opposing counsel without the opportunity to consult AI systems. Those situations, he argued, require independent legal reasoning that cannot be outsourced to technology.

AI Is Still Part of The Curriculum

Although the policy limits AI use during foundational coursework, the University of Chicago is not eliminating AI from legal education. Instead, the school plans to introduce the technology in stages.

Students enrolled in the mandatory first-year legal research and writing course will first learn to write legal memoranda and other documents without AI assistance before later incorporating AI tools into the drafting process.

The approach is intended to ensure students understand legal writing principles before using AI to enhance efficiency.

Beyond the first year, the law school has expanded AI instruction through specialized upper-level courses and established an AI laboratory where students can develop and experiment with legal technology tools.

Faculty members teaching elective courses will also retain flexibility to establish their own AI policies, although the laptop ban and in-class examination requirements will remain the default position unless professors choose otherwise.

Part of A Wider Debate In Legal Education

The University of Chicago’s decision comes as law schools across the United States grapple with the rapid emergence of generative AI. Earlier this year, the University of California, Berkeley School of Law introduced strict limits on student use of AI in academic work.

Berkeley’s policy sparked debate within the legal profession, with some critics arguing that it restricts legitimate educational uses of AI at a time when law firms increasingly expect graduates to be proficient with the technology.

Major law firms have invested heavily in AI-powered legal research, contract review and document analysis platforms, making technological literacy a valuable skill for new lawyers.

At the same time, judges, legal academics and bar associations have expressed concerns about AI-generated legal errors, fabricated case citations and overreliance on automated systems that may produce inaccurate or incomplete legal analysis.

Several courts have already introduced rules requiring lawyers to verify AI-generated legal filings or disclose when AI has been used in preparing court documents.

Chilton acknowledged that reactions among incoming students are likely to be mixed, particularly because laptops have become standard tools in university classrooms.

However, he said previous experiments in which individual professors prohibited laptops had generally received positive feedback after students experienced fewer distractions and greater classroom engagement.

Goldman Sachs has prohibited its employees from trading prediction market contracts tied to companies, elections, financial markets and geopolitical events, becoming one of the highest-profile financial institutions to tighten internal rules as regulators increase scrutiny of the rapidly growing industry.

According to Bloomberg, the investment bank recently updated its personal trading policy to bar employees from wagering on event contracts involving specific companies, electoral outcomes, macroeconomic data, financial market performance, and geopolitical developments.

The move comes amid mounting concerns that prediction markets, which allow users to bet on the likelihood of future events, could create new avenues for insider trading, particularly for employees with access to confidential corporate or market-sensitive information.

Goldman said staff who repeatedly violate the policy could face disciplinary action, including dismissal or the closure of their trading accounts. The bank also reserved the right to recover profits earned from prohibited trades.

Under the revised policy, if a trade is determined to be improper, Goldman can claw back any gains exceeding $200 or require that the amount be donated to a charitable organization.

When contacted by CNBC, a Goldman Sachs spokesperson declined to comment on the specifics of the revised policy but reiterated the firm’s longstanding position on insider trading.

The spokesperson noted that trading using material, nonpublic information is prohibited across every market in which Goldman operates.

Broad Range of Contracts Now Prohibited

The restrictions extend well beyond traditional securities trading. Employees are now barred from participating in contracts that speculate on whether Goldman itself will announce a corporate restructuring during a particular quarter or pursue a merger or acquisition.

The policy also prohibits trading on contracts linked to ceasefire timelines in ongoing armed conflicts, the future price of Bitcoin, regulatory approval of pending mergers and acquisitions, and other events where employees could possess privileged information unavailable to the public.

However, Goldman has stopped short of banning prediction market activity altogether. Employees remain free to participate in contracts related to sports and entertainment, where insider information is considered less likely to intersect with the firm’s business activities or client relationships.

Banks Adopt Different Approaches

Wall Street firms have been taking varied approaches to regulating employee participation in prediction markets as the platforms gain popularity.

According to Barron’s, JPMorgan Chase has adopted a more measured stance, advising employees to exercise caution before trading contracts related to the financial sector rather than imposing a blanket prohibition.

Meanwhile, hedge funds Point72 Asset Management and Balyasny Asset Management have introduced even stricter rules than Goldman by banning all prediction market trading in employees’ personal accounts, regardless of the subject matter.

The differing policies are seen as a pointer to the regulatory uncertainty surrounding an industry that has expanded rapidly over the past two years as prediction markets have evolved from niche platforms into venues attracting billions of dollars in trading volume.

Industry Faces Growing Regulatory Scrutiny

Goldman’s policy update follows what authorities have described as the first insider trading case involving a prediction market linked to a private-sector company.

In May, the Commodity Futures Trading Commission (CFTC) and the U.S. Department of Justice charged Michele Spagnuolo, a Google employee, with allegedly using confidential company information to profit from contracts traded on Polymarket.

According to the CFTC’s complaint, Spagnuolo, who allegedly traded under the username “AlphaRaccoon,” used inside knowledge relating to Google’s annual “Year in Search” rankings to place successful wagers.

Federal regulators alleged the activity generated approximately $1.2 million in profits.

The case marked a significant milestone because it demonstrated that insider trading laws can extend beyond conventional stock and options markets to prediction contracts tied to corporate events.

More Firms Expected To Tighten Rules

Bloomberg reported that among 50 companies surveyed, only three had formal policies governing employee participation in prediction markets, while two others said they were actively considering introducing restrictions.

Morgan Stanley said it already addresses the issue through its employee code of conduct.

At Bank of America, internal communications outlining new restrictions were reportedly being distributed to employees, according to a person familiar with the matter.

The issue has also reached the U.S. government.

In March, the White House warned officials against using nonpublic government information to trade prediction market contracts after unusual activity in futures markets preceded President Donald Trump’s public announcement that U.S. strikes against Iran would be paused.

The warning underscored growing concerns that government officials, like corporate insiders, could potentially exploit confidential information through prediction market platforms.

Together, Goldman’s tougher stance represents a notable shift given Chief Executive David Solomon’s previously positive comments about the emerging industry.

As recently as January, Solomon described prediction market platforms as “super interesting”. He also disclosed that he had met with the leaders of the two dominant companies operating in the sector.

The new policy suggests that while Goldman continues to recognize the commercial potential of prediction markets, it views the compliance risks associated with employee participation as high.

As prediction markets expand into areas covering corporate earnings, mergers and acquisitions, economic data releases, elections, and geopolitical events, financial institutions are facing a new challenge of ensuring that employees with access to confidential information do not use those platforms to gain an unfair trading advantage.

The latest policy update indicates that, for major Wall Street firms, prediction markets are no longer viewed solely as innovative forecasting tools. They are increasingly being treated as regulated financial markets that require many of the same compliance safeguards, surveillance measures and insider trading controls that already govern stocks, bonds and derivatives.

Traders on prediction market platform Kalshi are increasingly betting that the U.S. Federal Reserve could raise interest rates again before the end of the year, underpinning persistent concerns about inflation even as policymakers remain divided over the appropriate path for monetary policy.

The latest pricing on Kalshi assigns a 54% probability that the Fed will deliver at least one interest rate increase this year. Although that is slightly lower than the 56% probability seen a day earlier, it still suggests that market participants view another rate hike as more likely than not.

The contract tracks when the next increase in the federal funds rate will occur, with outcomes covering a hike before the end of 2026, before July 2027, or before the end of 2027.

Beyond this year, traders see an even greater likelihood that U.S. borrowing costs will move higher.

Kalshi currently assigns a 62% probability that the Federal Reserve will raise rates before July 2027 and nearly an 80% chance that a rate increase will occur by 2028. The market expectations emerged shortly after the Federal Reserve released minutes from its June policy meeting, offering investors a clearer picture of how deeply divided officials remain over the direction of interest rates.

According to the minutes, policymakers expressed sharply different views on where the federal funds rate should stand by the end of the year. The document noted that “many participants” believed the appropriate policy rate would remain within or slightly below its current target range by year-end. However, the minutes also revealed that “many other participants” concluded the appropriate level of interest rates would be above the current target range before the end of the year.

The split underscores the difficult balancing act confronting the central bank as it attempts to contain inflation without unnecessarily slowing economic growth. The Federal Reserve’s benchmark interest rate currently stands at 3.50% to 3.75%, where it has remained since December 2025.

The debate over future policy comes as inflationary pressures continue to challenge policymakers. The Fed’s preferred inflation measure, the Personal Consumption Expenditures (PCE) Price Index, rose 4.1% year over year in May, its highest annual reading since April 2023.

The stronger-than-expected inflation data have bolstered concerns that price pressures remain too persistent for the central bank to begin easing monetary policy. In addition to domestic inflation, policymakers are also monitoring geopolitical developments, including rising tensions in the Middle East, which have the potential to push energy prices higher and add further pressure to consumer prices.

Those factors have complicated the Fed’s inflation outlook and contributed to differing opinions among officials about whether policy should remain restrictive or become even tighter.

Kalshi traders are also signaling that they expect little prospect of lower interest rates this year. A separate prediction market asking how many rate cuts the Federal Reserve will implement during 2026 currently assigns approximately a 76% probability that there will be no rate cuts before year-end.

Those expectations have remained relatively stable in recent weeks. The probability of no rate cuts rose sharply from 68% to 77% on June 16, coinciding with the first Federal Open Market Committee meeting chaired by Federal Reserve Chairman Kevin Warsh.

Since then, the market’s expectations have changed only modestly, including after the release of the June meeting minutes.

The prediction market’s outcome will ultimately be determined by the Federal Reserve’s official policy decisions. The divergence between policymakers’ views reflects the uncertainty surrounding the U.S. economy. Some officials believe inflation is likely to moderate sufficiently to allow interest rates to remain unchanged or edge lower, while others argue that persistent price pressures warrant maintaining restrictive policy for longer or even tightening further.

However, analysts believe the June minutes have offered investors few clear signals about the Fed’s next move, but confirmed that monetary policy remains highly data-dependent.

Future inflation reports, labor market data, and broader economic indicators are expected to play a decisive role in determining whether the central bank holds rates steady or concludes that further tightening is necessary to return inflation to its long-term target.

With prediction markets assigning only a slim majority to another rate increase while policymakers themselves remain divided, expectations for the remainder of the year remain finely balanced.

Artificial intelligence companies continue to consume vast amounts of online content while sending relatively little traffic back to the websites that produce it, according to new data from Cloudflare.

The development has bolstered growing concerns over the sustainability of the internet’s long-standing economic model.

The latest figures, covering the week of July 1 to July 7, show that Anthropic remains the largest outlier among major AI developers, with its web crawlers requesting pages roughly 2,800 times for every single referral its services sent back to publishers.

While that marks a substantial improvement from early April, when Cloudflare estimated Anthropic’s crawl-to-referral ratio at around 8,800-to-1, the broader trend suggests the company’s AI systems continue to collect online content at a pace that far exceeds the amount of traffic they return.

Cloudflare’s data also indicates that the improvement may not represent a sustained change.

During the first week of May, Anthropic’s bots reportedly crawled websites approximately 24,700 times for every referral, illustrating how dramatically the ratio can fluctuate over relatively short periods.

Among the major AI companies tracked by Cloudflare, OpenAI ranked second behind Anthropic, followed by AI search startup Perplexity. Microsoft and Google recorded lower crawl-to-referral ratios, while search engine DuckDuckGo stood out as one of the few companies approaching a more balanced exchange.

According to Cloudflare, DuckDuckGo generated roughly one referral for every three webpage crawls, a ratio far closer to the traditional relationship that has historically existed between search engines and publishers.

The data comes as AI-powered search tools increasingly replace conventional web searches.

For decades, website owners generally accepted search engines crawling their content because indexing generated traffic that could be monetized through advertising, subscriptions, or product sales.

The arrangement created a mutually beneficial ecosystem.

Search engines gained access to information, while publishers received visitors who generated revenue.

Generative AI is fundamentally changing that relationship.

Instead of directing users to external websites, AI chatbots and AI-powered search engines increasingly provide complete answers within their own interfaces. As a result, users often obtain the information they need without ever visiting the original source.

That shift has become one of the publishing industry’s biggest concerns as AI adoption accelerates. This is because if AI systems continue extracting content while reducing referral traffic, publishers could face declining advertising revenue and fewer incentives to invest in producing original journalism, research and educational material.

Cloudflare’s crawl-to-referral metric has therefore become an increasingly watched indicator of how AI companies interact with the broader web. Rather than measuring the absolute amount of data collected, the metric compares how frequently AI crawlers request webpages against how often users are sent back to those same sites.

Although the ratio does not capture every aspect of publisher traffic, it offers a proxy for whether AI companies are maintaining the economic exchange that has historically supported the open internet.

Anthropic’s position is particularly notable because the company has frequently emphasized AI safety and responsible development as core elements of its business strategy. The latest figures have therefore renewed debate over what constitutes responsible behavior when AI systems rely heavily on publicly available web content for search, retrieval and other services.

At the same time, Anthropic has recently taken a firm stance against the unauthorized use of its own AI technology. The company has criticized competitors for allegedly using outputs generated by Anthropic’s models to improve their own systems through a practice commonly referred to as model distillation. Anthropic argues that such activities violate its terms of service because they involve using its AI-generated outputs to train competing models.

Many companies in the AI industry now object to competitors using their proprietary AI outputs for commercial purposes while simultaneously relying on vast quantities of online content created by publishers, writers, researchers and other content producers.

Publishers have argued that AI companies should compensate content creators when their material contributes to commercial AI products, particularly as chatbot-generated answers reduce visits to original websites. Anthropic is currently facing a lawsuit for unlawful use of publishers contents.

Anthropic has previously challenged Cloudflare’s analysis.

The company said it could not independently verify Cloudflare’s methodology for calculating crawl-to-referral ratios and argued that recently introduced search features are increasing the number of users directed back to publishers. That suggests that Anthropic believes the reported figures may not fully reflect how its services generate referral traffic.