Alan Marling is taking to the streets to shine a critical light on Twitter.
The 34-year-old artist and writer, who has more than 16,000 followers on the social network, spent over a dozen evenings this year with a projector and colored lights, beaming his commentaries onto the outside walls of Twitter's San Francisco headquarters.
Twitter "validates Nazis," said one of his recent messages, which the California-based Marling says are meant to protest Twitter's "hypocrisy." The projection came complete with a blue checkmark like the ones the social network uses to verify high-profile users.
Another message urged Twitter to "Ban @realDonaldTrump" and argued that the US president uses the service to "create divisiveness and conflict." Last week, Marling projected the message "Trump's Dog Whistle," with an arrow pointing to the main Twitter sign on the building's facade. And after Trump tweeted a series of controversial anti-Muslim videos, Marling projected a message demanding that Twitter CEO Jack Dorsey resign.
Marling's work -- about which Twitter declined to comment -- is just one example of the responses to messages seen on the social network this year.
From fears about Russia manipulating US public opinion, to the #MeToo hashtag encouraging women from all walks of life to speak up about sexual assaults, Twitter has been at the center of memes and movements that shook society in 2017.
Revelations that millions of messages are being generated by bots and fake accounts have prompted many of us to question what we're reading. And as the preferred platform for President Donald Trump, Twitter has become a facilitator of the nation's divisive chatter.
"Twitter can put you on an emotional roller coaster," said Jeff Hancock, director of Stanford University's Social Media Lab. "Social media in general has changed how we process and disseminate information. It's just a matter of if you believe everything you see and read."
Hancock said our growing distrust of Twitter is being amplified by fake news as well asused social networks to meddle in the 2016 US presidential election.
In short, this nonstop drumbeat of negativity has an effect.
"We begin to have a natural suspicion" of Twitter, Hancock said. "While we've always been mistrustful about new technology, there's more in the media now about misinformation [and] that brings a heightened awareness."
On Dec. 18, Twitter began enforcing new rules to reduce the amount of abusive and hateful content on the platform. Those include prohibitions against users' names or bios promoting violence and hate; potentially suspending accounts permanently that threaten violence or serious physical harm and death; and a ban on accounts featuring hate symbols and images.
However, the company said its efforts will be a work in progress. "We'll evaluate and iterate on these changes in the coming days and weeks," according to a blog post.
Here's just some of what's happened on Twitter in the last year.
#MeToo sparks a global movement
On Oct. 15, activists, actors and women from across the US began tweeting with a two-word hashtag, #MeToo. They were inspired by sexual harassment allegations against Hollywood mogul Harvey Weinstein published in detail in The New York Times and The New Yorker. Within 24 hours, the hashtag was trending globally and had been tweeted nearly a half-million times. So far, tweets referring to the hashtag have been seen more than 15 billion times, according to social media analyzer Brandwatch.
The tweeting started with actress Alyssa Milano, who encouraged millions of people to speak up about the abuse they've suffered. The movement forced many to grapple with the prevalence of sexual harassment and abuse.
The #MeToo movement so quickly and so dramatically changed the national conversation that Time magazine named "the Silence Breakers" its Person of the Year.
"#MeToo is proving there's no longer anyone who's untouchable and protected when it comes to sexual harassment," said Jennifer Grygiel, a Syracuse University professor of communications. "Women are now realizing they have the power to speak out without fear, to be strong and supportive of each other."
From Hollywood to Silicon Valley to Capitol Hill, virtually no industry has been left untouched. Oscar winner Kevin Spacey and Pixar boss John Lasseter; tech evangelist Robert Scoble; and politicians Al Franken, John Conyers and Roy Moore have all been chastised for alleged sexual misconduct. Morning show hosts Charlie Rose and Matt Lauer weren't spared, either.
Still, there are many more women whose stories have yet to be told.
"What about the daily harassment the clerical worker goes through? What about the cleaning lady, and the female bus driver who won't do 'favors'?" asked economist and activist Julianne Malveaux. "I feel pain for those women whose stories still aren't being told."
#MeToo is now beyond hashtag activism. It's led to protests and speeches globally, and cost some prominent men their livelihood.
From Russia, with(out) love
During last year's presidential election, many Twitter users unknowingly felt the influence of Russian bots, or automated software posing as real users.
For example, Twitter counted 36,000 fake accounts between September and November 2016 that generated 1.4 million automated election-related tweets and received 288 million views. One study, released in March, said bots make up about 48 million Twitter accounts, or 15 percent of the social network's monthly users. Twitter, meanwhile, maintains that about 27 million of its accounts are bots.
Not all these bots are bad. Some simply publish articles from news sites or post the latest weather. But spreading misinformation and inflammatory content is the aim of others, such as the hundreds of Russia-linked bots that stoked both sides of the NFL national anthem debate, using the hashtags #standforouranthem and #takeaknee.
It used to be you could tell from stilted phrasing that a computer program was behind a tweet. No longer. In the past year, bot masters have taught their programs to sound more human. Now they're finding that using humans and bots together can spread a message even more effectively.
"That makes it really hard to tell if an account is a bot and then therefore take it down," Samantha Bradshaw, an Oxford University researcher, told PBS' Frontline last month. And that means once again you can't always trust what you're reading on Twitter.
More frustratingly, it's "highly unlikely" we'll be completely free from seeing malicious bots on Twitter, said Adam Sharp, the company's former head of news, government and elections. Twitter offers its users a choice to be completely anonymous, making it hard to trace who's behind a tweet.
"It may end up being a constant arms race, as these foreign actors have already gained a bit of an edge," Sharp said.
Our new political landscape
One thing that marked the 2016 election was how much toxic online behavior found its way into modern politics. During that time, trolls from the darkest part of the internet came together with political activists and extremists to spread conspiracy theories, disinformation and false news across social media.
At the center of much of it is President Trump, who had said before being sworn in on Jan. 20 that he planned to cut back on his tweeting habits. Yet since then, he's turned to Twitter to attack NFL football players, North Korea, the FBI and anyone else he happens to be feuding with.
The message has become clear: Politics will now be fought and won on social media.
"Twitter in particular, is the new political landscape," said longtime Republican political strategist Rick Wilson, who's also a Trump critic. "If you're not fighting in social media, you're not fighting effectively."
That mindset was on full display when Democrats in the US Senate took to live video streams to protest what they said was a rushed process for voting on new tax legislation.
It's also part of why Marling continues his protests outside Twitter's headquarters. On the night of Nov. 29, he projected the hashtag #complicit just above the Twitter logo.
The word, he said, is "a sign of our times."
First published Dec. 8, 6 a.m. PT.
Update, Dec. 18 11:10 a.m. PT: Adds details about Twitter enforcing it updates to reduce abusive and hateful content on the social network.
Special Reports: CNET's in-depth features in one place.
Solving for XX: The tech industry seeks to overcome outdated ideas about "women in tech."