The race to lead in AI is spurring a fresh wave of corporate acquisitions and investments as tech companies seek to show the world they aren't sleeping on the red-hot technology, Axios' Hope King and I report.

Why it matters: AI is poised to reshape many industries, and the pressure is on CEOs to prove they have an AI strategy.

On the acquisition front, Databricks announced this week it is paying $1.3 billion (or $21 million per employee) to buy two-year-old startup MosaicML, which specializes in open source AI models.

• Thomson Reuters, meanwhile, said Tuesday it will pay $650 million for legal AI assistant Casetext.

Hundreds of millions more is being invested by large tech companies for minority stakes in AI startups.

• Earlier this month, Salesforce Ventures said it was doubling the size of its generative AI fund — to $500 million.
• Venture arms from Google, Microsoft and others are also actively investing in AI startups.
• Corporate investing can help larger firms get an insider view of startups' technical work and spot potential acquisition targets.

It's not just tech companies that are opening up their checkbooks. Consulting and other firms outside of tech are also stepping up their AI investments.

• Accenture last week closed its purchase of Bangalore, India-based Flutura, which it says will boost its ability to offer AI services in the energy, chemicals, metals, mining and pharmaceutical industries.
• PwC US recently announced it's planning to invest $1 billion over the next three years in AI projects and the upskilling of its 65,000-person workforce.

Yes, but: The old Latin saying caveat emptor (buyer beware) applies here, too. Everyone these days says their company is an AI company.

Be smart: Not everyone needs to build their own ChatGPT.

• For many companies, success will mean taking other companies' AI technology and combining it with their unique expertise or data sets.
• At the same time, many existing businesses will find themselves under pressure from nimble new rivals who have an AI-based approach that can operate cheaper, better or faster.

Between the lines: Among early stage investors, there are now two different worlds — generative AI startups and then everything else, PitchBook has found.

• "Investors are measuring generative AI and all other startups on very different criteria: the first on sheer potential and the latter on cold, hard revenue," PitchBook notes.
• Median pre-money valuations for early-stage rounds of generative AI companies have jumped 16% so far this year compared to last year, while prices for all other startups raising Series A or Series B have dropped by nearly 24%, according to PitchBook data.

More than half of Americans say they have experienced hate or harassment online, according to a new survey from the Anti-Defamation League, with a dramatic rise in incidents over the last 12 months, especially among teens.

Why it matters: Experts say what happens online is causing significant real-world harm and also keeping large numbers of people from fully participating in an increasingly digital society.

By the numbers: The increases in harassment were seen in almost all demographics but were particularly pronounced among underrepresented religious and ethnic groups as well as LGBTQ+ people.

• The highest level of online hate and harassment reported was among transgender Americans, with three quarters saying they had been harassed at some point and more than half saying they had been targeted in the past year.
• Nearly half of others in the LGBQ+ community (47%) said they had experienced online hate or harassment in the past year.
• Rates were also high among Blacks and Muslims, with 38% of both groups reporting being the subject of attacks over the last 12 months. More than a quarter of Jewish respondents (26%) said they were targeted in the past 12 months, up from 21% in the 2022 survey.
• More than half of teens (51%) reported some form of harassment in the last 12 months, compared to 36% in the prior year's survey. A third of adults, meanwhile, experienced some form of online hate or harassment, up from 23% a year ago.

Between the lines: Some platforms are safer than others. More than half of those who reported being harassed said they were targeted on Facebook. Meanwhile, the number of people reporting harassment on Twitter and Reddit was up significantly from the prior year.

What they're saying: "Online hate and harassment isn't just 'hurtful speech,'" Yael Eisenstat, head of the ADL's Center for Technology and Society, said in a statement to Axios. "It pushes people out of the conversation, impinging on their own freedom of expression, causing emotional distress, reputational and economic harm."

• It's also inspiring real-world violence, Eisenstat said. "Mass shooters are finding inspiration in the misogynist, anti-LGBTQ+ and antisemitic content coursing through their feeds. And social media posts on mainstream platforms that target trans people have been directly linked to bomb and death threats against hospitals that provide gender-affirming care."

The big picture: ADL is among a number of civil liberties groups urging social media companies to do more. GLAAD recently gave failing grades to all the major players for their efforts to combat online hate, noting they often fail to enforce their own policies.

• As Axios reported, Facebook is still not fully reviewing all reports of rule violations, a practice it instituted during the pandemic.

Methodology: The survey of 2,139 individuals was conducted online from March 7 through March 24, 2023 by YouGov.

The U.S. is considering fresh restrictions on exports to China of AI chips from Nvidia and others, according to a new Wall Street Journal report.

Why it matters: The U.S. and China are locked in a complicated relationship when it comes to tech. Both countries have imposed roadblocks on the other in a quest for superiority, while the two remain dependent on one another.

Details: The Commerce Department could impose the restrictions as soon as next month, the Journal said, citing people familiar with the matter.

• Under the new rules, shipments of AI chips would require a license before being sold to customers in China and certain other countries.
• The new rules reportedly go further than those imposed by the Biden administration last year.

Between the lines: The U.S. has already restricted exports of the most advanced AI chips, including processors from AMD, Nvidia and others, amid a broader effort by the U.S. to hobble China's ability to advance its own chipmaking sector.

On Tap

• Axios Codebook author Sam Sabin is moderating a session at Collision on how to turn bots from a security liability into an advantage.
• MWC Shanghai, the Chinese edition of the GSMA's cell phone trade show runs through Friday.

ICYMI

• The Wall Street Journal reports that up to 80% of video ads sold by Google aren't meeting promises, a charge the company disputes.
• The Supreme Court ruled Tuesday that to be a "true threat" a cyberstalker must have recklessly disregarded that their words or actions will be perceived as a threat. (Fast Company)
• TikTok is killing TikTok Now, its BeReal clone, after less than a year. (TechCrunch)

If you liked the cool fake shadow art I shared on Monday, check out this kid creating a frog-like shadow.